nopperl/arxiv-image-text · Datasets at Hugging Face

dedup-isc-ft-v107-score float64 0.32 1	uid stringlengths 32 32	text stringlengths 0 32.5k	paper_id stringlengths 1 14	original_image_filename stringlengths 5 222
0.492273	438fa9256bb44b3f8d1e9c048dffe674	Toy model example of a set of 2D constraints, where the 1D projections hide the discrepancy between the two data sets. The darker and lighter shade correspond to the 68% and 95% confidence regions respectively.	2012.09554	figure_1.jpg
0.489467	d180f98d00164de188fc3726120bffd2	68 % and 95 % confidence regions of the constraint on the differences in parameters as measured by DES and , constructed as discussed in <ref>. The markers indicate the location of the synthetic input shifts. The corresponding a-priori Gaussian tension is shown in <ref>.	2012.09554	figure_3.jpg
0.365842	6dacf06e98a040ac94883496dc80b885	Example of the prior-volume dependence of R. In amber and red are two gaussians that are at a 3 σ tension. The black dotted line is the prior (note that it is not normalized, to make it easier to visualize). When we use a uniform prior in the range [-10, 10] (left panel), R is much smaller than one, which means the data sets are in tension. When we increase the prior to [-200, 200] (right panel), R becomes greater than one, indicating agreement. This example, although extreme, illustrates a possible issue of the Bayes ratio as a tension metric.	2012.09554	figure_4.jpg
0.441661	c31dc1fd586c47b0b8038d1ef52677a5	The fractional Fisher information on cosmological parameters for computed using the KL modes from its update with simulated DES. Each line shows the fractional contribution of each KL mode to the total information on a given parameter. The sum of values in each row is one. The numbers on top of the figure show the fractional error improvement of DES over Planck for each KL mode.	2012.09554	figure_5.jpg
0.376208	a1a6dc286526478cb25e38f14e9caab9	A graphical illustration of the main results of <ref>. Different points show the tension calculated by each tension metric as a function of the input shifts. The error bars in the green points correspond to sampling errors, which can be calculated for evidence-based methods by re-sampling the nested sampling weights.	2012.09554	figure_6.jpg
0.374268	e9310e36829947ce8fa8b6608bb0b394	Tension estimates given by different metrics versus the corresponding Bayes ratio. Shaded regions highlight Jeffreys' scale used to interpret the Bayes ratio, with the vertical line separating “Tension" to the left and “Agreement" to the right.	2012.09554	figure_7.jpg
0.458165	ac570aa34caf416aa7e7e2341d08d95c	68 % and 95 % confidence regions of the joint marginalized posterior probability distributions for Dark Energy Survey Year 1 Cosmic Shear, 3× 2pt and 5× 2pt likelihoods, and for the 2015 TTTEEE likelihood.	2012.09554	figure_9.jpg
0.475116	8a0a0c33698e4d6ba2450cffbb6e2338	Taxonomy of Smartphone Sensing Studies for the Well-being of Young Adults from a Human Science Perspective	2012.09559	SmartphoneSensingV18.jpg
0.471036	4e5e7fda36ce42e988fee69592256b52	The excitation energies E_x(I^+_1), the energy ratios R_I+2 (where R_I ≡ E_x(I^+_1) / E_x(2^+_1)), and the B(E2;I → I-2) values for 6 protons and 6 neutrons in the pf shell with the schematic interaction H(x)= x ( ∑_j_αε_j_α n_j_α + g V_P ) + κ V_Q [see Eq. (<ref>)]. Panel (a) is for I=2 and panel (b) for I=4. SM is the abbreviation for the shell model; SDG_ GS is for the SDG-pair approximation with the generalized seniority-like approach approach; and SDG_ CG is for the SDG-pair approximation with the conjugate gradient approach.	2012.09560	fig1.jpg
0.411794	72e195ddc4484021974002e782e3d9c3	The yrast states of the N=26 isotones, ^46Ca, ^48Ti, ^50Cr, and ^52Fe. The NPA calculations are performed in the SDG_ CG⊕ SDG_ GS space. The experimental data are taken from <cit.>.	2012.09560	fig2.jpg
0.404549	3eb682a3876f403cbd510a41d7c78e8b	The yrast states of ^44-48Ti. SDG_ HF is the abbreviation for the SDG-pair approximation with the Hartree-Fock approach, and PHF is for the angular momentum projected Hartree-Fock calculation. The experimental data are taken from <cit.>.	2012.09560	fig3.jpg
0.385321	13634a690f964873902faa7d9aa1c5dc	The yrast states of ^48Cr, ^50Cr, and ^52Fe. The experimental data are taken from <cit.>.	2012.09560	fig4.jpg
0.361619	d075c8b6969f4116a72bf4204319d992	The yrast states of ^60-64Zn. The experimental data are taken from <cit.>.	2012.09560	fig5.jpg
0.367306	2e22e420784e4d83b5f89f709de6210f	The yrast band of ^64,66Ge and ^84Mo. The experimental data are taken from <cit.>.	2012.09560	fig6.jpg
0.378524	ee6fbd0d740a418db289f489d9efc935	The yrast band of ^108-112Xe. The experimental data are taken from <cit.>.	2012.09560	fig7.jpg
0.423212	d67afcf9f53a444d9e45d6a8689f6c39	The yrast band of ^112,114Ba.	2012.09560	fig8.jpg
0.420381	9bcb9fb42d4145c4a2f59a49b2cf022c	Mean of mixed-Hamming error rates on SNAP ego-networks for Mixed-SLIM_appro based on different choice of T, where T is in {1,2, …, 20}. x-axis: T, y-axis: mean of mixed-Hamming error rates.	2012.09561	mixedSLIMapproTSNAP.jpg
0.47131	8ca804e9278a4682847524e10ba9a91f	Number errors on six empirical networks for Mixed-SLIM_appro based on different choice of T, where T is in {1,2, …, 20}. x-axis: T, y-axis: number errors.	2012.09561	mixedSLIMapproTreal6.jpg
0.461634	e1d51c5de0964317ae81b6e5f898c772	Mean of mixed-Hamming error rates on SNAP ego-networks for Mixed-SLIM_τ when τ=c d̅, where c is in {0,0.1,0.2,…,2}. x-axis: c, y-axis: mean of mixed-Hamming error rates .	2012.09561	mixedSLIMtauCSNAP.jpg
0.431725	9436e3328b214bf08779e227e39484bf	Number errors on the six empirical datasets for Mixed-SLIM_τ when τ=c d̅, where c is in {0,0.1,0.2,…,2}. x-axis: c, y-axis: number errors.	2012.09561	mixedSLIMtauCreal6.jpg
0.504051	1cff271a97f546128631c5d24ff234e6	Diagrams contributing to the one loop corrections for the process γ^(p_1) γ^(p_2) → l(p_3) l̅(p_4)	2012.09565	4pt_loop.jpg
0.46128	553c1b0c33154c65a91d9ea602bf37f5	TCS (left panel) and BH (right panel) processes. The blob on the proton line denotes the nucleon structure, whereas the blob on the lepton line denotes the QED amplitude, including radiative corrections.	2012.09565	BH_process_real.jpg
0.611202	a003972216d54735acc2ea651489500f	Born contribution (upper panel) and s-channel Δ-pole contribution (lower panel) to the Compton amplitude. While for the Born contribution only the sum of s and u- channel diagrams is gauge invariant, the s-channel Δ-pole contribution is gauge invariant by itself.	2012.09565	Born_dvcs.jpg
0.45994	8241f51a527847e3915cd2e23e4425f0	One-loop QED correction to the vertex.	2012.09565	Vertex1L.jpg
0.476008	98991b3139f84162a503b66061373129	The γ^* N Δ vertex Γ_γ N Δ^βμ (left diagram) and its adjoint, Γ̃_γ N Δ^αν (right diagram).	2012.09565	delta_vertex.jpg
0.584101	7454efd58a2c4ec5878edb2cbcb06d6c	Diagram representing the doubly virtual Compton process.	2012.09565	dvcs_blob.jpg
0.596817	27dd7afc5fcc47afbf45e9e2ffe64827	Handbag diagrams for the doubly virtual Compton amplitude. The single (composite) lines represent quarks (nucleons) respectively. The blobs represent the GPDs.	2012.09565	handbag.jpg
0.531438	88880df2d0a3403abfd055a5fee4b30c	Vacuum polarization diagram. The fermion loop can be either electrons or muons.	2012.09565	vacuumpol.jpg
0.496992	5de5ae79a745482d8e5d3824345744ff	Testing the two-point connectivity prediction (<ref>) with h =0 level sets of the 2D GFF on periodic square lattices of size L× L. Main. The averaged two-point connectivity P_2(x,y) as a function of ln\|x-y\|/ln L. The dots are numerical measure with L = 32, 64, …, 2048; see inset (a) for color code. The solid line represents the prediction (<ref>) (with the exact slope 1/π and arbitrarily adjusted offset C). Inset (a) The slope as a function of system size, obtained by a global linear fit of each curve. A convergence to the prediction 1/π as L→∞ is observed. Inset (b) The raw data. Plotting P_2 against \|x-y\| as usual, we observe no collapse. The dotted lines are a guide to the eye.	2012.09570	2p.jpg
0.471103	078bb59d8c9e407689412ee1c2ec6314	(a) A loop model configuration on a honeycomb lattice and the associated height field ϕ (heights are in unit of π). The color of a loop (blue/green) determines the increment of ϕ across it; ϕ is constant in regions between loops. (b) A point z encircled by d_z = 4 loops, and a path from the boundary to z (without crossing other loops that are not drawn). The field along the paths gives a random walk realization {Z_j} = 0, -1, 0, 1. (c) Two points x and y satisfying q = 2, d_x = 3, d_y = 4, connected to the boundary by a branching path. The random walk realization {X_j } = 0,-1,0,-1 and {Y_j } = 0, -1, 0, 1,0 share the first q steps.	2012.09570	loops.jpg
0.452367	3bb3746a291e4d06b70668bbb8bfbfc1	Percolation threshold of 2D GFF level sets on square lattices of size L × L. Main: The existence probability of a percolating cluster P_L(f), where f = (level set area) / L^2 is the level set's mean density. Inset: Finite-size scaling collapse according to the correlation length prediction (<ref>), identifying ξ = L. The critical value f_c = 1/2 regardless of lattice details. All numerics in this paper are done with real-valued 2D GFFs on periodic square lattices, generated by the standard Fourier filter method <cit.>.	2012.09570	pc.jpg
0.416418	d0b5769e98a64810ae30cb8faa90763d	Level set percolation in random fields with different roughness exponents. (a) When H = -1 < 0, there is a percolation transition, with fractal clusters appearing at criticality. (b) When H = 0 (2D GFF), the critical clusters are logarithmic fractals. (c) When H = 1/2 > 0, there is no sharp transition, and the clusters are not fractal.	2012.09570	three.jpg
0.394157	256d1e024fc14155be3acbe642109bc2	Samples of 4 of the 1000 identities present in the COX Face database used to perform the experiments.	2012.09571	cox_samples.jpg
0.482066	e130e652a98842b2a723ba3f2151bdcf	Pipeline of OSDe-SVM.	2012.09571	pipeline.jpg
0.455586	292ac2296a924b4cb3e39712a17ce188	Open-Set Dynamic Ensembles of SVM (OSDe-SVM) is able to incorporate new knowledge and correcting wrong updates by adding and removing classifiers in an unsupervised way. The system is designed to work under open-set recognition conditions.	2012.09571	pipeline_general.jpg
0.478761	4c9799f8f9ee44409987164f8134ef92	Pipeline of OSDe-SVM when self-healing is performed.	2012.09571	pipeline_selfhealing.jpg
0.589298	f045b271ce2d4ea383d3d7a3a938628a	Adaptation steps performed during the experiments. INI stands for initialisation, UP for update and SH for self-healing. The last step corresponds to the beginning of the second lap.	2012.09571	updates.jpg
0.425139	6b1e6158eb794b20a0323a5c54d6315d	Phase diagram for filament in MP assay under elastic loading in N-ω̃ plane, with k̃_m = 450, ṽ_0 = 24.24, f̃_s = 60, f̃_d = 19.2, K̃_T = 83 kept fixed. The lines denote the linear stability phase boundaries between linearly stable (s), stable spiral (ss), unstable spiral (us), and linearly unstable (u) phases. The points denoted by and indicate decaying oscillations and limit cycle oscillations, respectively, corresponding to the full non-linear dynamics in Eq.(<ref>).	2012.09572	N_W_pd2.jpg
0.383193	26998c20c3a74807a5f8cf1537b9b479	The linear stability phase boundary between the stable spiral (blue: ss) and unstable spiral (yellow: us) phase in the plane of elastic loading stiffness K_T and MP number N is shown using the heat map of C-AB in Eq.(<ref>). The color box shows the mapping for the values of the function. Parameters used correspond to kinesin-microtubule assay, keeping k̃_m = 450, ω̃ = 20, f̃_s = 60, f̃_d=19.2, ṽ_0=24.24 fixed. The points denoted by and indicate decaying oscillations (ss) and stable limit cycle oscillations (us), respectively, corresponding to the full non-linear dynamics shown in Eq.(<ref>). Here we express K_T in units of pN/nm.	2012.09572	NvsKT.jpg
0.448306	a0f2687bcc5f43b195449650d41c2e57	(color online) Kymographs show the time evolution of the probability distributions of MPs with extension ỹ: (a) in the attached state (P_a), and (b) in the detached state (P_d). The color-box describes the values of probability distributions. The relaxation dynamics are determined by dimensionless diffusion constants D̃_a=15.6, D̃_d = 17.4, and the relaxation rate ν̃= 80. Other parameter values used are ṽ_0 = 50, f̃_d = 4, k̃_m = 4.57, ω̃ = 5, f̃_s = 14.29, N=160 and K̃_T = 4.57.	2012.09572	distbn.jpg
0.445432	b3ef0552a2e4421192169e0be6bb134a	(color online) Schematic diagram of the model where a motile MT filament is attached with a harmonic trap of stiffness constant K_T. When attached, kinesin walks along the filament towards right with a velocity v_m, pulling the filament towards left.	2012.09572	fig1.jpg
0.443724	5fc1cfce8d744dc7a203d339171f1060	(color online) Phase diagram under constant loading in ṽ_0-f̃_̃d̃ plane shown using the heat map of the quantity q defined in Eq.(<ref>). We use parameter values typical of a microtubule- kinesin system, k̃_m = 450 (k_m=1.7pN/nm), f̃_s = 60, ω̃ = 20 and use N = 5 number of MPs. The color-box shows the mapping of q-values to the color code. The light blue (yellow) region denotes unstable (stable) phase under perturbation.	2012.09572	phdia_const_load.jpg
0.471421	1ec99490bdbf49d9801945cc7cfea22d	(color online) Time evolutions of (a) microtubule displacement x̃(), and (b) mean kinsein extension ỹ(). (c) A parametric plot of x̃() and ỹ() shows stable limit cycle. (d) Correlation function of microtubule displacement C_x̃() in the time-periodic steady state. The red (green) lines in all these figures correspond to numerical simulations (solutions to the Fokker-Planck based mean field equations). We used ṽ_0 = 50, f̃_d = 4. All other parameter values are the same as in Fig. <ref>.	2012.09572	relaxation_oscil.jpg
0.514583	28c93461ec614fe2b09466e46784809f	(color online) Phase diagram for harmonically trapped microtubule-kinesin assay in ṽ_0-f̃_d plane, at fixed k̃_m = 4.57, ω̃ = 5, f̃_s = 14.29, N=160 and K̃_T = 4.57. The points denote the two phases characterized by decaying oscillations () and stable limit cycle oscillations () obtained from numerical simulations. The solid black line identifies the boundary of supercritical Hopf-bifurcation predicted by Eq.(<ref>).	2012.09572	simu_phdia.jpg
0.45998	8ae5d2a87aca486d8cca10256ff6c85a	Phase diagram for filament in MP assay under harmonic trap in ṽ_0-f̃_d plane, keeping k̃_m = 450, ω̃ = 20, f̃_s = 60, N = 5 and K̃_T = 83 fixed. The phases and phase boundaries are indicated in the same manner as in Fig.<ref>.	2012.09572	v0_fd_pd1.jpg
0.464271	af14046634c04c4a8ba1db5f5926dcc9	Trajectories of three scenarios 𝒮_1, 𝒮_2, 𝒮_3 (from the STMS dataset described in Section <ref>) are displayed on the left. A batch ℬ composed of trajectories sampled from the three scenarios is represented on the right.	2012.09573	batch.jpg
0.475696	77176e857f7c409fb683f8c723399547	Plots of the empirical distribution of the q_i descriptors and the fitted distributions. They are computed from the 10,000 normal trajectories of the STMS validation dataset. Each scenario is processed separately to compute the q_i descriptors.	2012.09573	distributions.jpg
0.408267	778dc3ac2a9b4beebbcc71c0faa4459f	Recurrent auto-encoder network computing the latent representation of trajectories. Numbers indicate the input dimension for each layer.	2012.09573	encoder-decoder-deeper-V2-horizontal.jpg
0.422147	11af929f2b814f72adfcce484948b983	Precision, recall and F-measure computed w.r.t. the salient trajectory class. These metrics quantify the trajectory saliency detection performance for different values of λ on the STMS validation set. λ values are sampled every 0.05 within [0,5].	2012.09573	evaluation_varying_lambda.jpg
0.42773	4159bb5146b0459e8c36c9037bacd9ca	Training scheme of the recurrent auto-encoder with the two associated losses.	2012.09573	metric-learning-unsupervised.jpg
0.427905	cc276ca21916446db24c926703973dda	Calculated enthalpy as function of pressure, of the three most stable YN_2 polymorphs, relative to the enthalpy of the ThC_2 structure. The dashed line marks the enthalpy of bulk YN (B1 structure) plus 12N_2 (ϵ-N_2 structure).	2012.09576	DeltaH_YN2_reazioni.jpg
0.46441	a09ecdfa000148ebb86d2ef2bff42d9f	The simplest Lewis structure with only σ bonds of the (N_4)^6- moiety found in the P2_1/c#2 structure. In reality the chemical bonding has a partial multicenter π contribution.	2012.09576	N4_6-.jpg
0.396703	845f6065f2ac45d1b602e8234e1a8132	Calculated volume per formula unit as function of pressure. The vertical lines indicate the transition pressures.	2012.09576	Volume_YN2_sections.jpg
0.418758	4987dbaf67154ddf8b3845e4f6f83dc7	Quadrilateral element with inner interface.	2012.09581	Q4_with_ED.jpg
0.383831	a301cf71444d46cdb2d4ea5c959abff9	Separation modes (α_1=α_n0, α_2=α_m0).	2012.09581	Q4_with_ED_Mode_I_II.jpg
0.398528	2383640cddda43ae91dc51c3e245dd21	Crack nucleation criterion.	2012.09581	Q4withED_Crit_For_Nucleation.jpg
0.39951	a965eaa1a51c45e9a8eabfbdd0047551	Four point bending test: Deformed finite element mesh (displacements magnified by a factor of 250).	2012.09581	Q6withED_BGP_4BP_DefMesh_scaled_250times_wholeModel.jpg
0.485916	f2f7c707530c4e6f84f1efe4aea751a3	Four point bending test calculated without incompatible modes (Q4 with ED): a) Nucleated cracks in the mesh and b) specific dissipated fracture energy at the end of the analysis..	2012.09581	Q6withED_BGP_4PB_Q4_cracks_vs_diss.jpg
0.378363	2493a218696343d1bcdca245329d1baf	Four point bending test: Total load versus relative displacement at the notch mouth.	2012.09581	Q6withED_BGP_4PB_Q4_vs_Q6_response_curve.jpg
0.446246	3d9ce57ef44347fab336d7220dbd4516	Four point bending test: a) Nucleated cracks in the mesh and b) specific dissipated fracture energy at the end of the analysis.	2012.09581	Q6withED_BGP_4PB_cracks_vs_diss.jpg
0.447407	e9b4a35bf0364754b32e671db6b74123	Four point bending test: Geometry, boundary and loading conditions.	2012.09581	Q6withED_BGP_4PB_geometry.jpg
0.477649	d4d26075ed4745358f2a97b5ae46321b	Four point bending test: Finite element mesh.	2012.09581	Q6withED_BGP_4PB_mesh.jpg
0.399256	3eb10533e7874ec998f194956cb16f6c	Model A	2012.09581	Q6withED_BGP_DENS_ModelA_geometry.jpg
0.424551	c4c1d242a82c415eb8c60342f74699cc	Model B	2012.09581	Q6withED_BGP_DENS_ModelB_geometry.jpg
0.52426	00686ef6bfab44b68aaa381e880c8b92	Comparison with another numerical simulation.	2012.09581	Q6withED_BGP_DENS_ResponseCurve_WuEtAl_2015_leg2.jpg
0.464023	1bba8be2a87849cdb73571108c93ead4	Comparison with experiment.	2012.09581	Q6withED_BGP_DENS_ResponseCurve_experiment_leg2.jpg
0.395208	fe8e7daddfdd4ab6ae969e8e54737ae3	Kalthoff's test: Considered meshes with nucleated cracks.	2012.09581	Q6withED_BGP_Kalthoff_CracksPlot_All.jpg
0.430903	96264f32d742499998a11c4378281fce	Kalthoff's test: Specific dissipated fracture energy (mesh with 19800 elements).	2012.09581	Q6withED_BGP_Kalthoff_DissPlot_FineMesh_All.jpg
0.444365	88e4bebd11d1497c91458e75ef8ced7e	Kalthoff's test: Geometry and boundary conditions.	2012.09581	Q6withED_BGP_Kalthoff_Geom_and_Load.jpg
0.429624	405186ed61194826bfadb0d97de11795	Kalthoff's test. Mesh discretization used for different numerical methods: Gradient method <cit.> ≈4.000.000 FEs, Phase field <cit.> ≈160.000 FEs, Strain injection technique <cit.> ≈15.000 FE, and novel Q6 with ED 3.280 FE.	2012.09581	Q6withED_BGP_Kalthoff_MeshDiscret.jpg
0.404259	36f5454683704c7d82c66709b446ca59	Kalthoffâs test: Comparing results of Q6ED (first row) with Q4ED (second row) for plain strain formulation and 3280 elements. The first column shows meshes with nucleated cracks and the last two columns display specific dissipated fracture energies due to cracking in Mode I and Mode II fashion.	2012.09581	Q6withED_BGP_Kalthoff_Q6_vs_Q4_All_Plot.jpg
0.505098	c86100218a1941078b39106146fda4c1	SENS test: a) Deformed mesh and b) nucleated cracks in the mesh at the end of simulation.	2012.09581	Q6withED_BGP_SENS_DefMesh_and_cracks.jpg
0.401599	da23e0053a7949bea47a91c5f2377e42	SENS test: Details of the mesh with the upper secondary crack path. Left: Nucleated cracks. Right: Specific dissipated fracture energy for Mode I.	2012.09581	Q6withED_BGP_SENS_DetailA.jpg
0.460156	be54936c9bb5463e9d20d5f54e3501d5	SENS test: Dissipation due to Mode I separation.	2012.09581	Q6withED_BGP_SENS_Dissipation_2.jpg
0.449764	23161c452ef6471baa81a863ba539899	SENS test: Geometry and boundary conditions.	2012.09581	Q6withED_BGP_SENS_Geom_and_Load.jpg
0.490882	99f5d34602f34abc82739c40a6ed90ff	SENS test: Specific dissipated fracture energy for a) Mode I and b) Mode II.	2012.09581	Q6withED_BGP_SENS_dissMI_II_Leg.jpg
0.521403	77031c8cee3d41758a59ad0bc13d3203	Tension test: Problem illustration.	2012.09581	Q6withED_BGP_Tens_II_Geom_and_Load.jpg
0.428142	38643be8ff964698bd09e09df710b944	Tension test: Dissipated fracture energy versus imposed displacement for mode I (left) and mode II (right).	2012.09581	Q6withED_BGP_Tens_II_GraphsDiss_p.jpg
0.395667	ec4c2b3f7e614d1290f600c56b876a66	Tension test: Reaction versus imposed displacement. Right: Detail A.	2012.09581	Q6withED_BGP_Tens_II_GraphsR_p.jpg
0.418382	c25c574a2e5f4a77866347a28ad99908	Tension test: Meshes with nucleated cracks, green elements have weakened strength.	2012.09581	Q6withED_BGP_Tens_II_cracksAll.jpg
0.381181	e1b5acb8c5f74dc88639d9474e61a03f	Tension test: Details of distorted meshes with nucleated cracks.	2012.09581	Q6withED_BGP_Tens_II_cracksPart.jpg
0.440142	1b74cce3a52b4acebf9192dc962bb1f8	Tension test: Element-wise presentation of dissipated fracture energy at λ=0.05.	2012.09581	Q6withED_BGP_Tens_II_dissAll2.jpg
0.45544	69bc0e755087425aaf89676ccb61c4ae	Uniaxial cyclic test. Left: Problem illustration. Right: Loading protocol with the load factor λ = λ( t ) as a function of the monotonically increasing quasi-time parameter t ≥ 0.	2012.09581	Q6withED_BGP_Tens_I_Geom_and_Load.jpg
0.591194	ed970ce451fd4894b369f9005a57db25	Uniaxial cyclic test. Left: Reaction versus imposed displacement. Right: Position of embedded discontinuity.	2012.09581	Q6withED_BGP_Tens_I_Response_and_crack.jpg
0.526819	82894ac0244643698805ab37536a68f0	Uniaxial cyclic test. Left: Traction versus jump in displacement. Right: Traction versus softening variable.	2012.09581	Q6withED_BGP_Tens_I_inter_response.jpg
0.458256	c571958aa559449ba9071032163a07f7	Thermal shock: Nucleated cracks (first row) and specific dissipated energy (second row) at τ=5 for three different intensities of the temperature drop (2, 4 and 8). The dissipated fracture energy is presented as a percentage of G_fn, denoted as η_Gf. The mesh size is 80 × 20.	2012.09581	Q6withED_BGP_ThermalShock_CrackedMesh.jpg
0.470827	2731f4786bff4e058a8007c400125cb8	Thermal shock. Left: Geometry and boundary conditions. Right: Error complementary functions (Erfc) distribution along the height.	2012.09581	Q6withED_BGP_ThermalShock_Geometry.jpg
0.433548	90dddbb873f74d63bd105c7ac9fa294f	Thermal shock: Dissipated fracture energy for different intensities of the temperature drop (a=0.9, 2, 4 and 8).	2012.09581	Q6withED_BGP_ThermalShock_GraphDiss.jpg
0.455965	dc64c95da5854d2886bcd4ba448ccdee	Thermal shock: The “zig-zag” softening function.	2012.09581	Q6withED_BGP_ThermalShock_ZigZagFunction.jpg
0.422874	0b53d3dd80fe48ac904ce358530fafd8	Kalthoff's test: Details of crack paths in the plain strain model with 19800 finite elements.	2012.09581	Q6withED_Kalthoff_Branching_PlaneStrain.jpg
0.489568	236c30d737814b08886ce36a9565be03	Rigid-damage exponential softening in normal direction.	2012.09581	Softening_Exp_Constitutive_Law.jpg
0.485109	9891d70852204c5db84e194d014c8901	S_eff - S_0 for (μ/σ_0 , T/σ_0) = (1.035 , 0.110) as a function of α, where the condensate is restricted according to σ(x_1) = αcos(6 π (x_1/a) /). We show results for both lattice discretizations W_2 = W'_2 and W_2 = W”_2 and three lattice spacings a. S_0 is a physically irrelevant a-dependent shift chosen such that S_eff\|_α = 0 is identical for all three lattice spacings.	2012.09588	S_eff_alpha.jpg
0.46859	0f93ad6bebd34196af4ae3d077a86c00	The condensate σ(x_1) corresponding to one of the local minima at (μ/σ_0,T/σ_0) = (0.60,0.176) and lattice spacing aσ_0 = 0.237. There is essentially no difference between the two lattice discretizations W_2 = W'_2 and W_2 = W”_2. The dotted line represents a cosine function with the same wavelength and amplitude.	2012.09588	local_minimum_hbp.jpg
0.440669	769604ebd7ba4d22bdbac392af13d762	Perceived understandability (PU) of the tasks for the easy group and the hard group (three tasks, range 0 (easiest) to 10 (hardest), combined range 0 (easiest) to 30 (hardest)).	2012.09590	boxplot-rq1.jpg
0.541045	631e80ae535f419fb9ec701e409d65c5	The {5,4} and {7,3} tilings of a hyMERA network, respectively. Every hyMERA network is comprised of a Schläfli number {p,q} tiling that describes the geometric relationship of the bulk to its boundary.	2012.09591	54-73-together.jpg
0.486972	159f3caf10414e1ebf23b588facea2a3	Subfigures a) and b) show the binary MERA and the ternary MERA implementations <cit.>. Top tensors are shown as tensors t in the figure, whereas the isometries and unitaries are identified by key.	2012.09591	MERA_classes.jpg
0.46591	ffcad108d7fe47bf856491d619f99a2b	Two examples of perfect tensor networks: the hexagonal state, as originally described in <cit.>, which describes a pure holographic state, and the pentagonal code (also referred to as the HyPeC in <cit.>), which contains a bulk logical index for encoding a [[5,1,3]] holographic quantum error correction code. The HyPeC itself spurred the development of several other related holographic tensor network codes <cit.>. It should be noted here that the Schläfli number convention used in the perfect tensor literature follows the scheme {q,p}, whereas for hyMERA, the convention follows the form {p,q}. As such, the hexagonal state shown on the left is embedded on a {4,6} manifold tessellation; the HyPeC is embedded on a {4,5} manifold tessellation.	2012.09591	bothHaPPY.jpg
0.399279	9d5746cdb23a4e2da539455134109964	Scaling-dimension spectra {Δ_i} obtained after 1000 trial runs of diagonalizing the average two-site descending superoperator 𝒟̅(ℛ_2) in the {5,4}-parameterization. We apply the relation Δ_i = -log_s(λ_i) for the tensor decompositions {Y,Q,R}, {Y,Q,T}, and {Y,Q,S}, as shown in Subfigures a), b), and c), respectively. These decompositions are described in Section <ref> and <cit.>. The results shown suggest that the multitensor constraints themselves do not seem to limit the space of possible scaling dimensions to such an extent that a minimal model CFT would not be simulable, and that proper selection of the tensors in the tensor decomposition set may yield the commensurate scaling dimensions.	2012.09591	combined_average_decomp.jpg
0.488183	9d309883434e4c809518446d24d23df0	a) shows a diagrammatic representation of the two newly-introduced tensors, S and T. In b)-d), several variants of hyperinvariant tensor networks are displayed using these antisymmetric variants; in particular, variants b) and c) were subjected to numerical studies, as shown in Figure <ref>. These variants show that construction of a hyMERA tensor decomposition need not even be composed of strictly unitary tensors, as was previously thought in <cit.>.	2012.09591	multi-generalize.jpg
0.455643	c8c619b05eea47cb83ffb1ebd50657eb	Subfigure a) shows the example particular tensor decomposition solution applied to the set {A,B} for each of the two parameterizations of hyMERA considered in this article. The requisite relations that the set {Y,Q,R} for multitensor-constraint adherence in a hyMERA network are displayed; We additionally require that the finite set of reduced-density matrices ρ(ℛ_n) must also exhibit nontrivial entanglement spectra. Subfigure b) shows the construction of the sets of tensors {A,B} and {U,W} from consituent tensors {Y,Q,R} in the {7,3} parameterization. Finally, Subfigure c) displays the construction of the sets of tensors {A,B} and {U,W} from consituent tensors {Q,R} in the {5,4} version.	2012.09591	multilinear.jpg

0.492273

438fa9256bb44b3f8d1e9c048dffe674

Toy model example of a set of 2D constraints, where the 1D projections hide the discrepancy between the two data sets. The darker and lighter shade correspond to the 68% and 95% confidence regions respectively.

2012.09554

figure_1.jpg

0.489467

d180f98d00164de188fc3726120bffd2

68 % and 95 % confidence regions of the constraint on the differences in parameters as measured by DES and , constructed as discussed in <ref>. The markers indicate the location of the synthetic input shifts. The corresponding a-priori Gaussian tension is shown in <ref>.

2012.09554

figure_3.jpg

0.365842

6dacf06e98a040ac94883496dc80b885

Example of the prior-volume dependence of R. In amber and red are two gaussians that are at a 3 σ tension. The black dotted line is the prior (note that it is not normalized, to make it easier to visualize). When we use a uniform prior in the range [-10, 10] (left panel), R is much smaller than one, which means the data sets are in tension. When we increase the prior to [-200, 200] (right panel), R becomes greater than one, indicating agreement. This example, although extreme, illustrates a possible issue of the Bayes ratio as a tension metric.

2012.09554

figure_4.jpg

0.441661

c31dc1fd586c47b0b8038d1ef52677a5

The fractional Fisher information on cosmological parameters for computed using the KL modes from its update with simulated DES. Each line shows the fractional contribution of each KL mode to the total information on a given parameter. The sum of values in each row is one. The numbers on top of the figure show the fractional error improvement of DES over Planck for each KL mode.

2012.09554

figure_5.jpg

0.376208

a1a6dc286526478cb25e38f14e9caab9

A graphical illustration of the main results of <ref>. Different points show the tension calculated by each tension metric as a function of the input shifts. The error bars in the green points correspond to sampling errors, which can be calculated for evidence-based methods by re-sampling the nested sampling weights.

2012.09554

figure_6.jpg

0.374268

e9310e36829947ce8fa8b6608bb0b394

Tension estimates given by different metrics versus the corresponding Bayes ratio. Shaded regions highlight Jeffreys' scale used to interpret the Bayes ratio, with the vertical line separating “Tension" to the left and “Agreement" to the right.

2012.09554

figure_7.jpg

0.458165

ac570aa34caf416aa7e7e2341d08d95c

68 % and 95 % confidence regions of the joint marginalized posterior probability distributions for Dark Energy Survey Year 1 Cosmic Shear, 3× 2pt and 5× 2pt likelihoods, and for the 2015 TTTEEE likelihood.

2012.09554

figure_9.jpg

0.475116

8a0a0c33698e4d6ba2450cffbb6e2338

Taxonomy of Smartphone Sensing Studies for the Well-being of Young Adults from a Human Science Perspective

2012.09559

SmartphoneSensingV18.jpg

0.471036

4e5e7fda36ce42e988fee69592256b52

The excitation energies E_x(I^+_1), the energy ratios R_I+2 (where R_I ≡ E_x(I^+_1) / E_x(2^+_1)), and the B(E2;I → I-2) values for 6 protons and 6 neutrons in the pf shell with the schematic interaction H(x)= x ( ∑_j_αε_j_α n_j_α + g V_P ) + κ V_Q [see Eq. (<ref>)]. Panel (a) is for I=2 and panel (b) for I=4. SM is the abbreviation for the shell model; SDG_ GS is for the SDG-pair approximation with the generalized seniority-like approach approach; and SDG_ CG is for the SDG-pair approximation with the conjugate gradient approach.

2012.09560

fig1.jpg

0.411794

72e195ddc4484021974002e782e3d9c3

The yrast states of the N=26 isotones, ^46Ca, ^48Ti, ^50Cr, and ^52Fe. The NPA calculations are performed in the SDG_ CG⊕ SDG_ GS space. The experimental data are taken from <cit.>.

2012.09560

fig2.jpg

0.404549

3eb682a3876f403cbd510a41d7c78e8b

The yrast states of ^44-48Ti. SDG_ HF is the abbreviation for the SDG-pair approximation with the Hartree-Fock approach, and PHF is for the angular momentum projected Hartree-Fock calculation. The experimental data are taken from <cit.>.

2012.09560

fig3.jpg

0.385321

13634a690f964873902faa7d9aa1c5dc

The yrast states of ^48Cr, ^50Cr, and ^52Fe. The experimental data are taken from <cit.>.

2012.09560

fig4.jpg

0.361619

d075c8b6969f4116a72bf4204319d992

The yrast states of ^60-64Zn. The experimental data are taken from <cit.>.

2012.09560

fig5.jpg

0.367306

2e22e420784e4d83b5f89f709de6210f

The yrast band of ^64,66Ge and ^84Mo. The experimental data are taken from <cit.>.

2012.09560

fig6.jpg

0.378524

ee6fbd0d740a418db289f489d9efc935

The yrast band of ^108-112Xe. The experimental data are taken from <cit.>.

2012.09560

fig7.jpg

0.423212

d67afcf9f53a444d9e45d6a8689f6c39

The yrast band of ^112,114Ba.

2012.09560

fig8.jpg

0.420381

9bcb9fb42d4145c4a2f59a49b2cf022c

Mean of mixed-Hamming error rates on SNAP ego-networks for Mixed-SLIM_appro based on different choice of T, where T is in {1,2, …, 20}. x-axis: T, y-axis: mean of mixed-Hamming error rates.

2012.09561

mixedSLIMapproTSNAP.jpg

0.47131

8ca804e9278a4682847524e10ba9a91f

Number errors on six empirical networks for Mixed-SLIM_appro based on different choice of T, where T is in {1,2, …, 20}. x-axis: T, y-axis: number errors.

2012.09561

mixedSLIMapproTreal6.jpg

0.461634

e1d51c5de0964317ae81b6e5f898c772

Mean of mixed-Hamming error rates on SNAP ego-networks for Mixed-SLIM_τ when τ=c d̅, where c is in {0,0.1,0.2,…,2}. x-axis: c, y-axis: mean of mixed-Hamming error rates .

2012.09561

mixedSLIMtauCSNAP.jpg

0.431725

9436e3328b214bf08779e227e39484bf

Number errors on the six empirical datasets for Mixed-SLIM_τ when τ=c d̅, where c is in {0,0.1,0.2,…,2}. x-axis: c, y-axis: number errors.

2012.09561

mixedSLIMtauCreal6.jpg

0.504051

1cff271a97f546128631c5d24ff234e6

Diagrams contributing to the one loop corrections for the process γ^*(p_1) γ^*(p_2) → l(p_3) l̅(p_4)

2012.09565

4pt_loop.jpg

0.46128

553c1b0c33154c65a91d9ea602bf37f5

TCS (left panel) and BH (right panel) processes. The blob on the proton line denotes the nucleon structure, whereas the blob on the lepton line denotes the QED amplitude, including radiative corrections.

2012.09565

BH_process_real.jpg

0.611202

a003972216d54735acc2ea651489500f

Born contribution (upper panel) and s-channel Δ-pole contribution (lower panel) to the Compton amplitude. While for the Born contribution only the sum of s and u- channel diagrams is gauge invariant, the s-channel Δ-pole contribution is gauge invariant by itself.

2012.09565

Born_dvcs.jpg

0.45994

8241f51a527847e3915cd2e23e4425f0

One-loop QED correction to the vertex.

2012.09565

Vertex1L.jpg

0.476008

98991b3139f84162a503b66061373129

The γ^* N Δ vertex Γ_γ N Δ^βμ (left diagram) and its adjoint, Γ̃_γ N Δ^αν (right diagram).

2012.09565

delta_vertex.jpg

0.584101

7454efd58a2c4ec5878edb2cbcb06d6c

Diagram representing the doubly virtual Compton process.

2012.09565

dvcs_blob.jpg

0.596817

27dd7afc5fcc47afbf45e9e2ffe64827

Handbag diagrams for the doubly virtual Compton amplitude. The single (composite) lines represent quarks (nucleons) respectively. The blobs represent the GPDs.

2012.09565

handbag.jpg

0.531438

88880df2d0a3403abfd055a5fee4b30c

Vacuum polarization diagram. The fermion loop can be either electrons or muons.

2012.09565

vacuumpol.jpg

0.496992

5de5ae79a745482d8e5d3824345744ff

Testing the two-point connectivity prediction (<ref>) with h =0 level sets of the 2D GFF on periodic square lattices of size L× L. Main. The averaged two-point connectivity P_2(x,y) as a function of ln|x-y|/ln L. The dots are numerical measure with L = 32, 64, …, 2048; see inset (a) for color code. The solid line represents the prediction (<ref>) (with the exact slope 1/π and arbitrarily adjusted offset C). Inset (a) The slope as a function of system size, obtained by a global linear fit of each curve. A convergence to the prediction 1/π as L→∞ is observed. Inset (b) The raw data. Plotting P_2 against |x-y| as usual, we observe no collapse. The dotted lines are a guide to the eye.

2012.09570

2p.jpg

0.471103

078bb59d8c9e407689412ee1c2ec6314

(a) A loop model configuration on a honeycomb lattice and the associated height field ϕ (heights are in unit of π). The color of a loop (blue/green) determines the increment of ϕ across it; ϕ is constant in regions between loops. (b) A point z encircled by d_z = 4 loops, and a path from the boundary to z (without crossing other loops that are not drawn). The field along the paths gives a random walk realization {Z_j} = 0, -1, 0, 1. (c) Two points x and y satisfying q = 2, d_x = 3, d_y = 4, connected to the boundary by a branching path. The random walk realization {X_j } = 0,-1,0,-1 and {Y_j } = 0, -1, 0, 1,0 share the first q steps.

2012.09570

loops.jpg

0.452367

3bb3746a291e4d06b70668bbb8bfbfc1

Percolation threshold of 2D GFF level sets on square lattices of size L × L. Main: The existence probability of a percolating cluster P_L(f), where f = (level set area) / L^2 is the level set's mean density. Inset: Finite-size scaling collapse according to the correlation length prediction (<ref>), identifying ξ = L. The critical value f_c = 1/2 regardless of lattice details. All numerics in this paper are done with real-valued 2D GFFs on periodic square lattices, generated by the standard Fourier filter method <cit.>.

2012.09570

pc.jpg

0.416418

d0b5769e98a64810ae30cb8faa90763d

Level set percolation in random fields with different roughness exponents. (a) When H = -1 < 0, there is a percolation transition, with fractal clusters appearing at criticality. (b) When H = 0 (2D GFF), the critical clusters are logarithmic fractals. (c) When H = 1/2 > 0, there is no sharp transition, and the clusters are not fractal.

2012.09570

three.jpg

0.394157

256d1e024fc14155be3acbe642109bc2

Samples of 4 of the 1000 identities present in the COX Face database used to perform the experiments.

2012.09571

cox_samples.jpg

0.482066

e130e652a98842b2a723ba3f2151bdcf

Pipeline of OSDe-SVM.

2012.09571

pipeline.jpg

0.455586

292ac2296a924b4cb3e39712a17ce188

Open-Set Dynamic Ensembles of SVM (OSDe-SVM) is able to incorporate new knowledge and correcting wrong updates by adding and removing classifiers in an unsupervised way. The system is designed to work under open-set recognition conditions.

2012.09571

pipeline_general.jpg

0.478761

4c9799f8f9ee44409987164f8134ef92

Pipeline of OSDe-SVM when self-healing is performed.

2012.09571

pipeline_selfhealing.jpg

0.589298

f045b271ce2d4ea383d3d7a3a938628a

Adaptation steps performed during the experiments. INI stands for initialisation, UP for update and SH for self-healing. The last step corresponds to the beginning of the second lap.

2012.09571

updates.jpg

0.425139

6b1e6158eb794b20a0323a5c54d6315d

Phase diagram for filament in MP assay under elastic loading in N-ω̃ plane, with k̃_m = 450, ṽ_0 = 24.24, f̃_s = 60, f̃_d = 19.2, K̃_T = 83 kept fixed. The lines denote the linear stability phase boundaries between linearly stable (s), stable spiral (ss), unstable spiral (us), and linearly unstable (u) phases. The points denoted by and indicate decaying oscillations and limit cycle oscillations, respectively, corresponding to the full non-linear dynamics in Eq.(<ref>).

2012.09572

N_W_pd2.jpg

0.383193

26998c20c3a74807a5f8cf1537b9b479

The linear stability phase boundary between the stable spiral (blue: ss) and unstable spiral (yellow: us) phase in the plane of elastic loading stiffness K_T and MP number N is shown using the heat map of C-AB in Eq.(<ref>). The color box shows the mapping for the values of the function. Parameters used correspond to kinesin-microtubule assay, keeping k̃_m = 450, ω̃ = 20, f̃_s = 60, f̃_d=19.2, ṽ_0=24.24 fixed. The points denoted by and indicate decaying oscillations (ss) and stable limit cycle oscillations (us), respectively, corresponding to the full non-linear dynamics shown in Eq.(<ref>). Here we express K_T in units of pN/nm.

2012.09572

NvsKT.jpg

0.448306

a0f2687bcc5f43b195449650d41c2e57

(color online) Kymographs show the time evolution of the probability distributions of MPs with extension ỹ: (a) in the attached state (P_a), and (b) in the detached state (P_d). The color-box describes the values of probability distributions. The relaxation dynamics are determined by dimensionless diffusion constants D̃_a=15.6, D̃_d = 17.4, and the relaxation rate ν̃= 80. Other parameter values used are ṽ_0 = 50, f̃_d = 4, k̃_m = 4.57, ω̃ = 5, f̃_s = 14.29, N=160 and K̃_T = 4.57.

2012.09572

distbn.jpg

0.445432

b3ef0552a2e4421192169e0be6bb134a

(color online) Schematic diagram of the model where a motile MT filament is attached with a harmonic trap of stiffness constant K_T. When attached, kinesin walks along the filament towards right with a velocity v_m, pulling the filament towards left.

2012.09572

fig1.jpg

0.443724

5fc1cfce8d744dc7a203d339171f1060

(color online) Phase diagram under constant loading in ṽ_0-f̃_̃d̃ plane shown using the heat map of the quantity q defined in Eq.(<ref>). We use parameter values typical of a microtubule- kinesin system, k̃_m = 450 (k_m=1.7pN/nm), f̃_s = 60, ω̃ = 20 and use N = 5 number of MPs. The color-box shows the mapping of q-values to the color code. The light blue (yellow) region denotes unstable (stable) phase under perturbation.

2012.09572

phdia_const_load.jpg

0.471421

1ec99490bdbf49d9801945cc7cfea22d

(color online) Time evolutions of (a) microtubule displacement x̃(), and (b) mean kinsein extension ỹ(). (c) A parametric plot of x̃() and ỹ() shows stable limit cycle. (d) Correlation function of microtubule displacement C_x̃() in the time-periodic steady state. The red (green) lines in all these figures correspond to numerical simulations (solutions to the Fokker-Planck based mean field equations). We used ṽ_0 = 50, f̃_d = 4. All other parameter values are the same as in Fig. <ref>.

2012.09572

relaxation_oscil.jpg

0.514583

28c93461ec614fe2b09466e46784809f

(color online) Phase diagram for harmonically trapped microtubule-kinesin assay in ṽ_0-f̃_d plane, at fixed k̃_m = 4.57, ω̃ = 5, f̃_s = 14.29, N=160 and K̃_T = 4.57. The points denote the two phases characterized by decaying oscillations () and stable limit cycle oscillations () obtained from numerical simulations. The solid black line identifies the boundary of supercritical Hopf-bifurcation predicted by Eq.(<ref>).

2012.09572

simu_phdia.jpg

0.45998

8ae5d2a87aca486d8cca10256ff6c85a

Phase diagram for filament in MP assay under harmonic trap in ṽ_0-f̃_d plane, keeping k̃_m = 450, ω̃ = 20, f̃_s = 60, N = 5 and K̃_T = 83 fixed. The phases and phase boundaries are indicated in the same manner as in Fig.<ref>.

2012.09572

v0_fd_pd1.jpg

0.464271

af14046634c04c4a8ba1db5f5926dcc9

Trajectories of three scenarios 𝒮_1, 𝒮_2, 𝒮_3 (from the STMS dataset described in Section <ref>) are displayed on the left. A batch ℬ composed of trajectories sampled from the three scenarios is represented on the right.

2012.09573

batch.jpg

0.475696

77176e857f7c409fb683f8c723399547

Plots of the empirical distribution of the q_i descriptors and the fitted distributions. They are computed from the 10,000 normal trajectories of the STMS validation dataset. Each scenario is processed separately to compute the q_i descriptors.

2012.09573

distributions.jpg

0.408267

778dc3ac2a9b4beebbcc71c0faa4459f

Recurrent auto-encoder network computing the latent representation of trajectories. Numbers indicate the input dimension for each layer.

2012.09573

encoder-decoder-deeper-V2-horizontal.jpg

0.422147

11af929f2b814f72adfcce484948b983

Precision, recall and F-measure computed w.r.t. the salient trajectory class. These metrics quantify the trajectory saliency detection performance for different values of λ on the STMS validation set. λ values are sampled every 0.05 within [0,5].

2012.09573

evaluation_varying_lambda.jpg

0.42773

4159bb5146b0459e8c36c9037bacd9ca

Training scheme of the recurrent auto-encoder with the two associated losses.

2012.09573

metric-learning-unsupervised.jpg

0.427905

cc276ca21916446db24c926703973dda

Calculated enthalpy as function of pressure, of the three most stable YN_2 polymorphs, relative to the enthalpy of the ThC_2 structure. The dashed line marks the enthalpy of bulk YN (B1 structure) plus 12N_2 (ϵ-N_2 structure).

2012.09576

DeltaH_YN2_reazioni.jpg

0.46441

a09ecdfa000148ebb86d2ef2bff42d9f

The simplest Lewis structure with only σ bonds of the (N_4)^6- moiety found in the P2_1/c#2 structure. In reality the chemical bonding has a partial multicenter π contribution.

2012.09576

N4_6-.jpg

0.396703

845f6065f2ac45d1b602e8234e1a8132

Calculated volume per formula unit as function of pressure. The vertical lines indicate the transition pressures.

2012.09576

Volume_YN2_sections.jpg

0.418758

4987dbaf67154ddf8b3845e4f6f83dc7

Quadrilateral element with inner interface.

2012.09581

Q4_with_ED.jpg

0.383831

a301cf71444d46cdb2d4ea5c959abff9

Separation modes (α_1=α_n0, α_2=α_m0).

2012.09581

Q4_with_ED_Mode_I_II.jpg

0.398528

2383640cddda43ae91dc51c3e245dd21

Crack nucleation criterion.

2012.09581

Q4withED_Crit_For_Nucleation.jpg

0.39951

a965eaa1a51c45e9a8eabfbdd0047551

Four point bending test: Deformed finite element mesh (displacements magnified by a factor of 250).

2012.09581

Q6withED_BGP_4BP_DefMesh_scaled_250times_wholeModel.jpg

0.485916

f2f7c707530c4e6f84f1efe4aea751a3

Four point bending test calculated without incompatible modes (Q4 with ED): a) Nucleated cracks in the mesh and b) specific dissipated fracture energy at the end of the analysis..

2012.09581

Q6withED_BGP_4PB_Q4_cracks_vs_diss.jpg

0.378363

2493a218696343d1bcdca245329d1baf

Four point bending test: Total load versus relative displacement at the notch mouth.

2012.09581

Q6withED_BGP_4PB_Q4_vs_Q6_response_curve.jpg

0.446246

3d9ce57ef44347fab336d7220dbd4516

Four point bending test: a) Nucleated cracks in the mesh and b) specific dissipated fracture energy at the end of the analysis.

2012.09581

Q6withED_BGP_4PB_cracks_vs_diss.jpg

0.447407

e9b4a35bf0364754b32e671db6b74123

Four point bending test: Geometry, boundary and loading conditions.

2012.09581

Q6withED_BGP_4PB_geometry.jpg

0.477649

d4d26075ed4745358f2a97b5ae46321b

Four point bending test: Finite element mesh.

2012.09581

Q6withED_BGP_4PB_mesh.jpg

0.399256

3eb10533e7874ec998f194956cb16f6c

Model A

2012.09581

Q6withED_BGP_DENS_ModelA_geometry.jpg

0.424551

c4c1d242a82c415eb8c60342f74699cc

Model B

2012.09581

Q6withED_BGP_DENS_ModelB_geometry.jpg

0.52426

00686ef6bfab44b68aaa381e880c8b92

Comparison with another numerical simulation.

2012.09581

Q6withED_BGP_DENS_ResponseCurve_WuEtAl_2015_leg2.jpg

0.464023

1bba8be2a87849cdb73571108c93ead4

Comparison with experiment.

2012.09581

Q6withED_BGP_DENS_ResponseCurve_experiment_leg2.jpg

0.395208

fe8e7daddfdd4ab6ae969e8e54737ae3

Kalthoff's test: Considered meshes with nucleated cracks.

2012.09581

Q6withED_BGP_Kalthoff_CracksPlot_All.jpg

0.430903

96264f32d742499998a11c4378281fce

Kalthoff's test: Specific dissipated fracture energy (mesh with 19800 elements).

2012.09581

Q6withED_BGP_Kalthoff_DissPlot_FineMesh_All.jpg

0.444365

88e4bebd11d1497c91458e75ef8ced7e

Kalthoff's test: Geometry and boundary conditions.

2012.09581

Q6withED_BGP_Kalthoff_Geom_and_Load.jpg

0.429624

405186ed61194826bfadb0d97de11795

Kalthoff's test. Mesh discretization used for different numerical methods: Gradient method <cit.> ≈4.000.000 FEs, Phase field <cit.> ≈160.000 FEs, Strain injection technique <cit.> ≈15.000 FE, and novel Q6 with ED 3.280 FE.

2012.09581

Q6withED_BGP_Kalthoff_MeshDiscret.jpg

0.404259

36f5454683704c7d82c66709b446ca59

Kalthoffâs test: Comparing results of Q6ED (first row) with Q4ED (second row) for plain strain formulation and 3280 elements. The first column shows meshes with nucleated cracks and the last two columns display specific dissipated fracture energies due to cracking in Mode I and Mode II fashion.

2012.09581

Q6withED_BGP_Kalthoff_Q6_vs_Q4_All_Plot.jpg

0.505098

c86100218a1941078b39106146fda4c1

SENS test: a) Deformed mesh and b) nucleated cracks in the mesh at the end of simulation.

2012.09581

Q6withED_BGP_SENS_DefMesh_and_cracks.jpg

0.401599

da23e0053a7949bea47a91c5f2377e42

SENS test: Details of the mesh with the upper secondary crack path. Left: Nucleated cracks. Right: Specific dissipated fracture energy for Mode I.

2012.09581

Q6withED_BGP_SENS_DetailA.jpg

0.460156

be54936c9bb5463e9d20d5f54e3501d5

SENS test: Dissipation due to Mode I separation.

2012.09581

Q6withED_BGP_SENS_Dissipation_2.jpg

0.449764

23161c452ef6471baa81a863ba539899

SENS test: Geometry and boundary conditions.

2012.09581

Q6withED_BGP_SENS_Geom_and_Load.jpg

0.490882

99f5d34602f34abc82739c40a6ed90ff

SENS test: Specific dissipated fracture energy for a) Mode I and b) Mode II.

2012.09581

Q6withED_BGP_SENS_dissMI_II_Leg.jpg

0.521403

77031c8cee3d41758a59ad0bc13d3203

Tension test: Problem illustration.

2012.09581

Q6withED_BGP_Tens_II_Geom_and_Load.jpg

0.428142

38643be8ff964698bd09e09df710b944

Tension test: Dissipated fracture energy versus imposed displacement for mode I (left) and mode II (right).

2012.09581

Q6withED_BGP_Tens_II_GraphsDiss_p.jpg

0.395667

ec4c2b3f7e614d1290f600c56b876a66

Tension test: Reaction versus imposed displacement. Right: Detail A.

2012.09581

Q6withED_BGP_Tens_II_GraphsR_p.jpg

0.418382

c25c574a2e5f4a77866347a28ad99908

Tension test: Meshes with nucleated cracks, green elements have weakened strength.

2012.09581

Q6withED_BGP_Tens_II_cracksAll.jpg

0.381181

e1b5acb8c5f74dc88639d9474e61a03f

Tension test: Details of distorted meshes with nucleated cracks.

2012.09581

Q6withED_BGP_Tens_II_cracksPart.jpg

0.440142

1b74cce3a52b4acebf9192dc962bb1f8

Tension test: Element-wise presentation of dissipated fracture energy at λ=0.05.

2012.09581

Q6withED_BGP_Tens_II_dissAll2.jpg

0.45544

69bc0e755087425aaf89676ccb61c4ae

Uniaxial cyclic test. Left: Problem illustration. Right: Loading protocol with the load factor λ = λ( t ) as a function of the monotonically increasing quasi-time parameter t ≥ 0.

2012.09581

Q6withED_BGP_Tens_I_Geom_and_Load.jpg

0.591194

ed970ce451fd4894b369f9005a57db25

Uniaxial cyclic test. Left: Reaction versus imposed displacement. Right: Position of embedded discontinuity.

2012.09581

Q6withED_BGP_Tens_I_Response_and_crack.jpg

0.526819

82894ac0244643698805ab37536a68f0

Uniaxial cyclic test. Left: Traction versus jump in displacement. Right: Traction versus softening variable.

2012.09581

Q6withED_BGP_Tens_I_inter_response.jpg

0.458256

c571958aa559449ba9071032163a07f7

Thermal shock: Nucleated cracks (first row) and specific dissipated energy (second row) at τ=5 for three different intensities of the temperature drop (2, 4 and 8). The dissipated fracture energy is presented as a percentage of G_fn, denoted as η_Gf. The mesh size is 80 × 20.

2012.09581

Q6withED_BGP_ThermalShock_CrackedMesh.jpg

0.470827

2731f4786bff4e058a8007c400125cb8

Thermal shock. Left: Geometry and boundary conditions. Right: Error complementary functions (Erfc) distribution along the height.

2012.09581

Q6withED_BGP_ThermalShock_Geometry.jpg

0.433548

90dddbb873f74d63bd105c7ac9fa294f

Thermal shock: Dissipated fracture energy for different intensities of the temperature drop (a=0.9, 2, 4 and 8).

2012.09581

Q6withED_BGP_ThermalShock_GraphDiss.jpg

0.455965

dc64c95da5854d2886bcd4ba448ccdee

Thermal shock: The “zig-zag” softening function.

2012.09581

Q6withED_BGP_ThermalShock_ZigZagFunction.jpg

0.422874

0b53d3dd80fe48ac904ce358530fafd8

Kalthoff's test: Details of crack paths in the plain strain model with 19800 finite elements.

2012.09581

Q6withED_Kalthoff_Branching_PlaneStrain.jpg

0.489568

236c30d737814b08886ce36a9565be03

Rigid-damage exponential softening in normal direction.

2012.09581

Softening_Exp_Constitutive_Law.jpg

0.485109

9891d70852204c5db84e194d014c8901

S_eff - S_0 for (μ/σ_0 , T/σ_0) = (1.035 , 0.110) as a function of α, where the condensate is restricted according to σ(x_1) = αcos(6 π (x_1/a) /). We show results for both lattice discretizations W_2 = W'_2 and W_2 = W”_2 and three lattice spacings a. S_0 is a physically irrelevant a-dependent shift chosen such that S_eff|_α = 0 is identical for all three lattice spacings.

2012.09588

S_eff_alpha.jpg

0.46859

0f93ad6bebd34196af4ae3d077a86c00

The condensate σ(x_1) corresponding to one of the local minima at (μ/σ_0,T/σ_0) = (0.60,0.176) and lattice spacing aσ_0 = 0.237. There is essentially no difference between the two lattice discretizations W_2 = W'_2 and W_2 = W”_2. The dotted line represents a cosine function with the same wavelength and amplitude.

2012.09588

local_minimum_hbp.jpg

0.440669

769604ebd7ba4d22bdbac392af13d762

Perceived understandability (PU) of the tasks for the easy group and the hard group (three tasks, range 0 (easiest) to 10 (hardest), combined range 0 (easiest) to 30 (hardest)).

2012.09590

boxplot-rq1.jpg

0.541045

631e80ae535f419fb9ec701e409d65c5

The {5,4} and {7,3} tilings of a hyMERA network, respectively. Every hyMERA network is comprised of a Schläfli number {p,q} tiling that describes the geometric relationship of the bulk to its boundary.

2012.09591

54-73-together.jpg

0.486972

159f3caf10414e1ebf23b588facea2a3

Subfigures a) and b) show the binary MERA and the ternary MERA implementations <cit.>. Top tensors are shown as tensors t in the figure, whereas the isometries and unitaries are identified by key.

2012.09591

MERA_classes.jpg

0.46591

ffcad108d7fe47bf856491d619f99a2b

Two examples of perfect tensor networks: the hexagonal state, as originally described in <cit.>, which describes a pure holographic state, and the pentagonal code (also referred to as the HyPeC in <cit.>), which contains a bulk logical index for encoding a [[5,1,3]] holographic quantum error correction code. The HyPeC itself spurred the development of several other related holographic tensor network codes <cit.>. It should be noted here that the Schläfli number convention used in the perfect tensor literature follows the scheme {q,p}, whereas for hyMERA, the convention follows the form {p,q}. As such, the hexagonal state shown on the left is embedded on a {4,6} manifold tessellation; the HyPeC is embedded on a {4,5} manifold tessellation.

2012.09591

bothHaPPY.jpg

0.399279

9d5746cdb23a4e2da539455134109964

Scaling-dimension spectra {Δ_i} obtained after 1000 trial runs of diagonalizing the average two-site descending superoperator 𝒟̅(ℛ_2) in the {5,4}-parameterization. We apply the relation Δ_i = -log_s(λ_i) for the tensor decompositions {Y,Q,R}, {Y,Q,T}, and {Y,Q,S}, as shown in Subfigures a), b), and c), respectively. These decompositions are described in Section <ref> and <cit.>. The results shown suggest that the multitensor constraints themselves do not seem to limit the space of possible scaling dimensions to such an extent that a minimal model CFT would not be simulable, and that proper selection of the tensors in the tensor decomposition set may yield the commensurate scaling dimensions.

2012.09591

combined_average_decomp.jpg

0.488183

9d309883434e4c809518446d24d23df0

a) shows a diagrammatic representation of the two newly-introduced tensors, S and T. In b)-d), several variants of hyperinvariant tensor networks are displayed using these antisymmetric variants; in particular, variants b) and c) were subjected to numerical studies, as shown in Figure <ref>. These variants show that construction of a hyMERA tensor decomposition need not even be composed of strictly unitary tensors, as was previously thought in <cit.>.

2012.09591

multi-generalize.jpg

0.455643

c8c619b05eea47cb83ffb1ebd50657eb

Subfigure a) shows the example particular tensor decomposition solution applied to the set {A,B} for each of the two parameterizations of hyMERA considered in this article. The requisite relations that the set {Y,Q,R} for multitensor-constraint adherence in a hyMERA network are displayed; We additionally require that the finite set of reduced-density matrices ρ(ℛ_n) must also exhibit nontrivial entanglement spectra. Subfigure b) shows the construction of the sets of tensors {A,B} and {U,W} from consituent tensors {Y,Q,R} in the {7,3} parameterization. Finally, Subfigure c) displays the construction of the sets of tensors {A,B} and {U,W} from consituent tensors {Q,R} in the {5,4} version.

2012.09591

multilinear.jpg