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
|
|---|---|---|---|---|
0.449607 | 82af288296b141b381c50ce0538a6b23 | Architecture comparison between baseline neural network and energy conservation neural network. In the figure, the blocks represent layers of neural network (NN). u and v represents the earthquake wave displacement and velocity respectively. du and dv represent the derivative of u and v with respect to time. | 2012.14334 | baseline_vs_ecnn.jpg |
0.56707 | ee4440b81a5c48c1a35773ec765e3d64 | Illustrative architecture of the proposed energy conservation neural network. | 2012.14334 | ecnn_arch.jpg |
0.496891 | a93aa4fea4e641d6b78cbbf2351f6730 | Damping effects in energy conservation neural network. In the figure, the system energy decreases with time due to the damping effects. In addition, the maximum displacement magnitude u_t also decreases with time. | 2012.14334 | ecnn_damping.jpg |
0.435541 | 7e74b06e165b478e825445d8cec67c3e | The relationship between the input energy E_I and the damping energy E_D in a dynamic system. The system energy becomes stable when the input energy E_I equals to the damping energy E_D. | 2012.14334 | ecnn_input_energy.jpg |
0.465411 | eab0d1bbce7541959fc3bc05d547868b | Computation graph of neural network. | 2012.14334 | ecnn_latent_space.jpg |
0.521584 | 2c34d88c2e4e431dbaef2d78388363f5 | Computation graph for multiple degrees of freedom using neural network. | 2012.14334 | ecnn_mdof_arch.jpg |
0.460634 | ef0f7740a5a344f9b911b02cdad2fe7e | Soil modeling: stress-strain response curve | 2012.14334 | soil_stress_strain.jpg |
0.451799 | ff5294f129e143358ffeae254c1ed15d | Test error over the entire training set at the end of each epoch. Wider dimensions tend to have smaller testing error. | 2012.14334 | stats_by_dim.jpg |
0.406657 | 16cf9a9a52484c9c86b4dc4e008fed40 | Test error over the entire training set at the end of each epoch. Testing error decreases with the increasing number of hidden layers. | 2012.14334 | stats_by_layer.jpg |
0.487474 | 5ca441e9c93142b5bb4d246a9383981a | The trade-off between accuracy and speed with increasing model depth. Lower is better for both axes. | 2012.14334 | stats_layer_time.jpg |
0.444351 | d1a0b6860fb64e62af9474b7a0efc203 | prediction. | 2012.14336 | trend-1.jpg |
0.392483 | e0e7217d46344883b814a484a3c67ba0 | Mass posterior means and 90% credible intervals for inputted m_β near zero. It is possible to distinguish the mass from zero for true m_β≳0.04 eV, with outliers characterized by large uncertainties δ_inst (energy broadening standard dev.). | 2012.14341 | 12-7B_7D_mass-90--hdi-post-intervals.jpg |
0.425958 | 76907cbd19b1444fb819e47d0da3a394 | Distribution of mass ordering results with respect to m_L and mass sensitivity (Δ t=2 yrs). The normal (inverted) ordering was reported when a 98.5% (85.5%) η credible interval excluded one ordering and was consistent with the other. | 2012.14341 | 2nu-2yr-update_MH-reporting_mL-sensitivity-vs-m_L_985-855.jpg |
0.426176 | 52195a7bf4874fa0badb0b6f43e9e097 | For 2 yrs of data assuming normal (dots) and inverted (triangles) orderings, posterior means and intervals on η as a function of potential sensitivity to m_L, defined as C.I. width. | 2012.14341 | 2nu-2yr-update_eta-90-post-vs-sensitivity.jpg |
0.426336 | b647102ae6334836bebdb4d1bbe3a446 | Neutrino mass posterior means and 90% credible intervals as a function of inputted m_β, for a one-neutrino model and the assumed experimental design. Interval widths (“sensitivities") decrease with m_β, asymptoting at ∼5 meV. | 2012.14341 | cenpa_mass-90-log-hdi-post-residuals.jpg |
0.478 | 8e60484d2c9242c7a09e6e394ae78f54 | Example pseudo-spectra overlaid on predicted curves (Eq. <ref> numerically convolved with a Gaussian) for normal and inverted orderings, with m_L=0 eV and a 2 yr runtime. Spectra are plotted as a function of the difference between reconstructed energy and the T endpoint. | 2012.14341 | mass-ordering-data-comparison-relativeK.jpg |
0.45403 | 9134aa3be85348d78a41bc9f150ae64c | For one pseudo-experiment, example posterior probability density plots and 2D-histograms (in both contour and scatter plot form) for parameters in the two-neutrino spectral model. Posteriors were obtained by analyzing data (Δ t =1 yr) that was generated assuming a normal mass ordering. | 2012.14341 | newfig6_2.jpg |
0.422086 | 3e7429fa635a4334978fa1717ce5f613 | Numerical estimation of the distribution P(E| ρ̂; Ĥ) in a seven-levels (d=7) system in which Ĥ has eigenvalues { -1.6,-1.2,-0.6,0,0.4,1.3,1.7 }, and ρ̂ has eigenvalues { 0.395,0.224,0.151,0.115,0.079,0.0020,0.013 }. The histogram plots the empirical distribution of E(Ûρ̂Û^†; Ĥ) for a sample of 10^5 unitary matrices U ∈ U(7), which are distributed uniformly according to the Haar measure. The blue line is the probability density function of the Gaussian distribution P_G^(μ,Σ^(2))(E) with the variance Σ^(2) given by (<ref>). In this example Σ^(2)≃ 0.02024 and η_Ĥ≃ 0.2317. | 2012.14342 | plot1.jpg |
0.540437 | 6f0192898ec145a494bf68de0a4e9b2a | An elementary cell of the ring chain graph | 2012.14344 | Figure01.jpg |
0.513743 | 4146dcb2af03451984bfe991de548d49 | The band edge crossings indicated by black dots for ℓ=1, ℓ_3=1/3π in dependence on ℓ_1>0. The red lines correspond to flat bands indicated in the first two bullet points above, independent of ℓ_1. On the other hand, the dots at k^2=4 and 16 correspond to the flat bands of the second bullet point independent of ℓ_3. | 2012.14344 | Figure02.jpg |
0.470061 | ae57a9966c904cb9a93bf6660d19ef84 | The positive spectrum for ℓ=1 and ℓ_1=3/2π | 2012.14344 | Figure03.jpg |
0.502269 | 2edb4d57e5654cc195fe65b986e8b2fa | The positive spectrum for ℓ=1 and ℓ_1=1/3π | 2012.14344 | Figure04.jpg |
0.479909 | 1a94a9597b714369a11e432ab2dd3bd4 | Band width leading term vs. band index (referring to bands around the points nπ/ℓ_1) for ℓ_1= ℓ_3 = 4/11π, ℓ_2=18/11π, and ℓ=1 in comparison with the gap width | 2012.14344 | Figure05.jpg |
0.507069 | 808ffcec8feb432b8142c5892530f7d7 | The leading term vs. band index for ℓ_1= π/5, ℓ_2=13π/7, ℓ_3=π/7, and ℓ=1, again in comparison with the gaps width | 2012.14344 | Figure06.jpg |
0.447298 | 80dae0e3a2c44c9d8faa23f4becb2109 | The leading term vs. band index for ℓ_1=1, ℓ_2=ℓ_3=π, and ℓ=1 | 2012.14344 | Figure07.jpg |
0.42923 | a1dab28219f0446b83d377c95fc622a7 | The leading term vs. band index for ℓ_1=1, ℓ_2/ℓ_3=1+√(5)/2, and ℓ=1 | 2012.14344 | Figure08.jpg |
0.421939 | 218a6441ac9948bf9285398fd2d51214 | The same as above but with ℓ_1=π | 2012.14344 | Figure09.jpg |
0.450629 | 813ca250ba5a4910b6c418ec2be463cc | The full band spectrum in the situation of Example <ref> with ℓ=1 | 2012.14344 | Figure10.jpg |
0.468774 | eba874bfbe564bc5a93a3f519d0f1501 | The full band spectrum in the situation of Example <ref> with ℓ_1=π | 2012.14344 | Figure11.jpg |
0.480844 | b13cd0cf70fb4709a5c4b0720dfa0b4a | The negative spectrum for ℓ=3/2 and ℓ_3=2/3π as a function of ℓ_1>0, the dot indicates the band crossing point. | 2012.14344 | Figure12.jpg |
0.387792 | 8a0acf69da1149f9ab68c1c8edee3267 | The positive spectrum for ℓ=1/3. | 2012.14344 | Figure13.jpg |
0.550178 | da25537a7b30474cb7ca71ef333df6b6 | The band dependence on ℓ_3 for ℓ=1/2. | 2012.14344 | Figure14.jpg |
0.450736 | 882fe17a94e84593a4e6cc25419d6c56 | The positive spectrum for ℓ=1. The red lines mark embedded flat bands, the dots the points where the gaps close. | 2012.14344 | Figure15.jpg |
0.52507 | 9f24804228b0480b9479e9090094f2e2 | The band dependence on ℓ_1 for ℓ=1/2. | 2012.14344 | Figure16.jpg |
0.469918 | 2886210f8ecc414d82519cf67f0c07c8 | Overview of the proposed pipeline. Refer to Sec. <ref> for details. | 2012.14345 | architecture.jpg |
0.462985 | 0b4d3cdfda974077b91abf71de495bb1 | Example images of the datasets used for this work: a) iCWT dataset; b) pois in TABLE-TOP dataset; c) white in TABLE-TOP dataset; d) HO-3D; e) YCB-Video training set, f) YCB-Video test set. | 2012.14345 | dataset.jpg |
0.505192 | d7700056a9bc4ca680b4281ef12dc2b5 | Example predictions on the TARGET-UNLABELED before model adaptation, selected by the for SS as positives (red boxes) and negatives (green boxes). | 2012.14345 | detected_image_square.jpg |
0.504859 | 7ac58aacfde24fa5aae1f9c7b83533b7 | mAP comparison of pool-based (red) and stream-based (green) AL strategies with varying query budgets for iCubWorld (a) and YCB (b) domains. | 2012.14345 | plots.jpg |
0.446262 | a39f54d9f1b64871aba43f9cae03405b | Diagnostic plot for Event 1. The rows show a series of 0.5^∘×0.5^∘ filtered maps corresponding to PA4 f150, PA5 f150, PA6 f150, PA4 f220, PA5 f090 and PA6 f090 from the bottom to the top, while the columns correspond to individual 3-day chunks of data for the 2017-2019 observing seasons, with those that don't have exposure near this object omitted (so there are implicit, variable-length gaps between each column). The maps are in units of signal-to-noise ratio, spanning from -8 (blue) to +8 (red). Time increases from left to right. It is clear that this region is quiet until the last map where a strong point source suddenly appears in all arrays that hit it. The columns are not equi-spaced, and there is a gap of 12 days between the preceding thumbnail maps and the ones in which the source is detected. The source is evident in the PA4 time line but the data do not pass our data quality cuts. The time for the last map spans from 2019-11-7 00:00:00 to 2019-11-10 00:00:00. Unfortunately, there are no observations of this region in 2019 after those shown in this figure. There is a several month long gap before any 2020 coverage of this region, and the 2020 data is not yet ready for analysis. | 2012.14347 | fig1.jpg |
0.473904 | 876411397fad4ce39c6f2f0789d7e76e | High-resolution light curve for Event 1 made by subdividing the ACT detector arrays into 4 sub-sections that hit the source at slightly different times. We observe a rapid rise from about 180 to 600 mJy at f090 and from 600 to 1100 mJy at f150 over the course of 8 minutes followed by what might be the beginning of a slower decay. We do not detect anything at this location before this, even when averaging four years of data. The gap between -4 and 0 minutes on the x-axis corresponds to the separation between two of our detector arrays. Unfortunately there are no observations of this region in 2019 after this. No f220 data survived our data cuts during the event. | 2012.14347 | light_curve_cand_011_nogain.jpg |
0.45798 | 97bbe939266d4acc94c596256abea07e | The light curve for Event 3 . The event starts with a sudden rise from below our detection limit to about 300 mJy in all three of ACT's frequency bands. It then decays roughly linearly to nearly undetectable over the course of 3 days, after which there is a new jump up to about 150 mJy followed by a new decay. There are large gaps in our coverage of these coordinates after this, but by day 12-15 the source is undetectable in f150 and f220 and barely detectable in f090. We do not detect it after this. In sky maps built from all our data (including the event itself) we detect an object at this location only in f090, at 3.5 σ. | 2012.14347 | light_curve_cand_018.jpg |
0.522527 | eb2327a4573e44b9a0595bbe4c9d97fb | Comparison of the characteristic luminosity with other bright mm-wave star flares from the literature. act-1/2/3: Event 1/2/3 from this paper. alma-1/2: Proxima Centauri <cit.> and AU Mic <cit.> flares measured with ALMA. bima-1: Flaring of GMR-A in the Orion Nebula <cit.>. eff-1, pdbi-1: V773 Tau by <cit.> and <cit.> respectively. ovma-1: Gem <cit.>. scuba2-1: JW 566 <cit.>. | 2012.14347 | mm_flares.jpg |
0.485744 | 7d3b5bac20d14c4bbd6387bad43a1af4 | The distance between the weights of a valid count-constraint point to the local minima followed by the application of 𝒫_𝒞 to it | 2012.14348 | moon_prop.jpg |
0.488946 | 28d2f591d8774f25b14bc131da90e347 | A variety of curves that minimizes the MSE under different quantile constraints, and regular least squares minimization. | 2012.14348 | quantiles.jpg |
0.46817 | cc640c49a6a5423a851f9716b93ee97d | Overview of the methodology of Roofpedia. | 2012.14349 | Wireframe.jpg |
0.441684 | b05261c676b64bb482e1b5e154c01891 | Geospatial engineering of the predicted result: tagging buildings with sustainable rooftops based on detected greenery and solar panels on them. | 2012.14349 | geospatial.jpg |
0.424904 | 7c4b050228c74b0d86efa659d63e4a9b | Illustration describing the project and its organisation in three parts. | 2012.14349 | graphical_r.jpg |
0.446617 | 93519c359ce04ea684447dfe3463cd2f | Examples of green roof model confusions. | 2012.14349 | green-roof-confusions.jpg |
0.462101 | b4c707ced2b545358c1b1cc45a3cfe75 | Labelling solar panels and rooftop greenery across cities. | 2012.14349 | labelling.jpg |
0.434334 | accee2f7af194a00a61ff92da8690c9e | Model prediction results compared with ground truth. | 2012.14349 | model.jpg |
0.393227 | 09078418dc5942ebbd449f2fafce9091 | A visualisation for understanding the spatial density of solar roofs (example for Singapore). | 2012.14349 | singapore.jpg |
0.414296 | ee5854c10d4146e18cccb681036f816c | Examples of solar roof model confusions. | 2012.14349 | solar-roof-confusions.jpg |
0.417479 | 56c91921a70b4d5cb8419a8399c48959 | Tiling of satellite images and labels. | 2012.14349 | tiling.jpg |
0.428058 | e236a08711484b75bc85363f9124d7e9 | Mapping buildings with sustainable rooftops (example for New York City). | 2012.14349 | vis.jpg |
0.389683 | 30c258ae7bd1460e9de1794cb9e72d90 | DeepBeam in a vehicular ad hoc scenario. (a) Vehicle 1 is communicating with vehicle 3, vehicle 2 with vehicle 4. Using DeepBeam, vehicles 1 and 4 infer the reciprocal aoa by passively eavesdropping ongoing data transmissions. (b) Vehicles 1 and 4 can steer the beam toward each other when they need to exchange data. | 2012.14350 | adhoc.jpg |
0.467627 | 778730d82ddc4c47b5e7c33aa378175b | Configuration of the room where the single-RF-chain dataset was collected, and position of the radios in the basic and diagonal configuration, and of the obstacle. | 2012.14350 | radio_ni_positions.jpg |
0.476197 | 667c02e959b94b24b0984fbaacd2d37b | Illustration of the Proposed HPConv. Local feature maps also used to extract global feature maps. denote the concatnation over channel dimension. | 2012.14360 | hierarchical_pyramidal_convolution.jpg |
0.42263 | 0a583875bb02431f8c718aa7c86ef9f3 | The block-diagram of the proposed model. Our model can be divided to three main parts: the frontend module, the backend module and the consensus module. The input tensor and its shape of each module are given below. Our main contributions are highlighted in yellow. | 2012.14360 | overview_of_our_approach.jpg |
0.470119 | d497822c682d45fdae262b5decd3c77a | Illustration of the PyConv. The illustration is adapted from <cit.>. Local and global feature maps are extracted from the input feature maps respectively. ⊛ denotes the convolution operation. | 2012.14360 | pyramidal_convolution.jpg |
0.466408 | ea3b8a59b9c640ccafa04a8130977eba | An example of a video sample annotated as "ABOUT". Only frames at time step T = 9∼ 19 are related with the word "ABOUT". | 2012.14360 | video_sample_of_ABOUT.jpg |
0.477529 | 4d96d15dd20c468d9776d70c67510b5a | Example of Dense Folding. When the extent of a DenseData matches the stride of a parent StreamData, the parent/child combination can be replaced with a single larger DenseData. | 2012.14363 | dense-folding.jpg |
0.394266 | 7800eebc932c410e904101c2bae70c1c | MPI_Pack performance for a variety of 2D objects described as a vector or subarray datatype. “Size” is the total object size, “count” is the number of objects packed, and “contiugous block size” is the number of contiguous bytes in each block of the object. The pitch of each contiguous block is 512. TEMPI provides fast handling regardless of MPI datatype construction or datatype count. | 2012.14363 | pack_2d.jpg |
0.500323 | d48f9e6147da412087bc70665b7ee3f7 | Example of stream elision. When a child has only a single element, it can be removed from the Type tree. | 2012.14363 | stream-elision.jpg |
0.464146 | 96ece099a1ad447cae80eb0728b26677 | Example of stream flattening. When the parent stride allows repeated children to maintain a fixed stride between their elements, the parent and child can be flattened into a single stream. | 2012.14363 | stream-flatten.jpg |
0.432946 | 2d9927b3b9b243cd99353e0dc115227d | A 3D object with extent E_0 × E_1 × E_2 floats in an allocation A_0 × A_1 × A_2 bytes, and the corresponding linearized memory layout. | 2012.14363 | stride.jpg |
0.442694 | e90973a424324e42b3325229b8cf91a7 | MPI_Send / MPI_Recv latency for the one-shot, device, model-based automatic selection, and Summit MPI baseline. The vast majority of the performance improvement comes from the datatype handling, before the one-shot or device method is selected. | 2012.14363 | summit_2d_latency.jpg |
0.462066 | 407d577fba114f9bbdb513a4501e68c3 | Normalized latency of the one-shot, device, and model-based selection based on the measured system parameters. The model-based automatic selection reliably chooses the faster method with minimal overhead. | 2012.14363 | summit_2d_normalized.jpg |
0.487185 | d9f5a05f393d4520b1b6c5edc981296b | Device pack. | 2012.14363 | summit_device_pack.jpg |
0.44326 | f9c4bdcb8ea34d1aa4d70fa538134286 | Device unpack. | 2012.14363 | summit_device_unpack.jpg |
0.379734 | 8c773ad9bd7546e4b83b6f3258531470 | Performance of phases of halo exchange. Alltoallv time is higher for more ranks and more nodes. Pack/Unpack time is constant, as the amount of data moved per rank is unchanged. | 2012.14363 | summit_halo_exchange_latency.jpg |
0.416364 | 5fd66d1c5a234d18aae20c879fa01d2b | Speedup of entire halo exchange. Speedup is lower for larger number of ranks as datatype handling is a smaller portion of the total time. | 2012.14363 | summit_halo_exchange_sp.jpg |
0.485902 | 15bfcb54cbec461e96f5f2de16a51d20 | Partial values of T_device, T_oneshot, and T_staged, (excluding pack time), using the values from (a). | 2012.14363 | summit_model_no_pack.jpg |
0.451214 | 1554ca795f8d48a895cf8f1e48289dbf | One-shot pack. | 2012.14363 | summit_oneshot_pack.jpg |
0.464021 | 2488cf781bfd4f1eb341214c2db70913 | One-shot unpack. | 2012.14363 | summit_oneshot_unpack.jpg |
0.499262 | e9b65c991eca4bfcb4d005f55fd489ee | Measurements of T_d2h, T_h2d, T_cpu-cpu, and T_gpu-gpu on Summit. | 2012.14363 | summit_raw.jpg |
0.400127 | e147551c7b4a47f39b73cc064ac5ac46 | The application source file (202) includes the MPI header file provided by the system (204). It is compiled (203) and linked with the system MPI implementation. When the binary (205) is executed, symbols are resolved and the MPI code from the system MPI library is executed. The TEMPI source files (207) are compiled (208) into a dynamic library (209) using the system MPI header. The application is compiled as normal except for TEMPI being inserted into the link order (203), or an unmodified application can be used with the LD_PRELOAD mechanism. When the application is executed, any symbols defined by the TEMPI library will be resolved there (209), allowing the TEMPI code to be executed. Any others will be resolved in the system implementation. | 2012.14363 | tempi.jpg |
0.419883 | 13ec524cf8504506a989cb5a1451bf12 | Three different MPI C fragments to generate the 3D object from Fig. <ref> with A_0=256, A_1=512, A_2=1024, E_0=100, E_1=13, and E_2=47. The right-hand side shows the corresponding Type IR after translation, with parent above child . Equivalent objects can be represented differently and require a later transformation pass. | 2012.14363 | translation.jpg |
0.44459 | c71f25e0a560410a99cb7eb02df467bc | Time for MPI derived datatype creation and commit time for a variety of 3D objects described with subarray (0-2), hvector of vector (3-5), hvector of hvector of vector (6-11), and subarray of vector (12-14). The “create” component uses MPI_Type* and MPI_Type_create* family of MPI APIs to describe the type. The “commit” component is how much time is consumed in MPI_Type_commit. The trimean of 30000 executions of each phase is reported. Create time is unchanged (TEMPI does nothing) and is reported for comparison. TEMPI's performance varies due to different required canonicalization operations. TEMPI slows commit time substantially, but it still has a negligible impact on application run time. | 2012.14363 | type_commit.jpg |
0.434479 | 523b9c7f7fc546f187d4a4a1421c8286 | Marangoni forces F̅_ M/π Pe and F̅_ M, tot/π Pe as a function of imposed velocity U̅ together with corresponding concentration profiles (in the z̅x̅-plane and in three dimensions) in the decoupled limit Pe≪U̅. All results are from numerical FEM solutions of the axisymmetric diffusion-advection equation in two-dimensional angular representation with ρ<R̅=30. In the advective regime U̅≫ 1 a concentration boundary layer develops [see Eq. (<ref>)]. | 2012.14364 | FUA3.jpg |
0.475461 | fb4b847b95f44fe09260d8c84f994e6c | Iterative three-dimensional FEM results for F̅_ M/π Pe (top) and F̅_ M,tot/π Pe (bottom) as a function of U̅ for Pe = 0-50 for a cubic system with -7 < y̅ <0, 0<x̅<7, -7<z̅<7. Insets show the slopes F̅_ M/U̅π Pe and F̅_ M,tot/U̅π Pe as a function of Pe calculated from the results for U̅=0.1. Artificial symmetry breaking from lattice irregularities/defects is prevented by averaging all measured quantities over two simulations with U and -U. Blue open circles are results for Pe=0 from FEM solutions to the axisymmetric diffusion-advection equation in two-dimensional angular representation with R̅=30. The slope in the linear response regime for U̅≪ 1 is reduced according to Eq. (<ref>). Results for U̅≫ 1 are essentially not affected by strong Marangoni flows Pe≫U̅. | 2012.14364 | FUPer7.jpg |
0.547811 | 1250c03e5066456e9a2e4e63a3c7da12 | Iterative three-dimensional FEM results for F̅_ M/π Pe (top) and F̅_ M,tot/π Pe (middle) as a function of U̅ for Pe = 0 and Biot numbers k̅=0-400 for a half-cylindrical system with ρ <15, x̅>0, -4 < y̅ <0, Blue dashed lines are results for k̅=0 from FEM solutions to the axisymmetric diffusion-advection equation in two-dimensional angular representation with R̅=30. Forces in the diffusive regime U̅≪ 1 are reduced according to Eq. (<ref>). Forces in the advective regime U̅≫ 1 are reduced according to Eq. (<ref>). Bottom row: FEM results for F̅_ M/π Pe and F̅_ M,tot/π Pe as a function of Biot number k̅ for U̅=10^-4,100 in comparison to scaling results in Eqs. (<ref>) and (<ref>). | 2012.14364 | FUkcylr15new.jpg |
0.420673 | 6cb9a3c05cfd4dcebfafbbac0e2ce758 | Average Nusselt number as a function of U̅ for constant flux and constant concentration boundary conditions in the decoupled limit Pe≪U̅. All results are from numerical FEM solutions of the axisymmetric diffusion-advection equation in two-dimensional angular representation with ρ<R̅=30. | 2012.14364 | NusseltU4.jpg |
0.496079 | 772a366dab8142e08e0bf91a87e5fcbe | Different theory results for Pe-U̅_ swim swimming relations in a double-logarithmic plot in comparison to experimental results on PEG-alginate swimmers (red lines, obtained from the data in Fig. <ref>) and camphor boats from Ref. <cit.> (black lines and dots; black line is time-dependent data, black dots are data at fixed times but for different radii a). PEG-alginate swimmers are described well by the high Reynolds number theory without evaporation (Biot number k̅=0); camphor boats are best described by a high Reynolds number theory with Biot number k̅≈ 550. The corresponding low Reynolds number theories give significantly lower swimming speeds. | 2012.14364 | Uswimexp.jpg |
0.467098 | abac31ab48214a5db5f126859dc5f9d1 | Synthesis of PEG-alginate swimmers by pipetting microliter amounts of PEG-alginate solution into cross-linker solution. Side view of a PEG-alginate swimmer showing its half-spherical shape. | 2012.14364 | fig2.jpg |
0.452285 | f86c0953219c4d7eb2dd9e2e59975deb | Swimming trajectory of two PEG-alginate swimmers in a cylindrical container. Color-coded is the sign of the curvature, blue/red trajectories curve clockwise/counter-clockwise. Swimmers prepared according to same protocol exhibit individually different curving behavior (left: mostly counter-clockwise, right: mostly clockwise). Reflections at walls are of different duration. | 2012.14364 | fig5.jpg |
0.393319 | 35e06e29baeb456a837c0a83297d7d1d | PIV measurements of a PEG-alginate swimmer (A) after dripping and (B) in motion. The velocity scale bar refers to the fluid velocities, which are also color-coded for velocity. The white dashed arrows indicate the direction of the particle motion. On the right we indicate the fluid motion in (B) featuring radial Marangoni flow (blue), combined with flows corresponding to two counter-rotating vortices created by particle motion (green), and downstream Karman-like vortices (red). | 2012.14364 | fig6_small.jpg |
0.450682 | 93332c489c4642a6900bb1bd46e3823a | Left: Mass of the PEG-alginate swimmers as a function of time with an exponential fit m(t) = m_∞ + m_0exp(-t/τ_m) (see text); error bars (light blue) denote the standard deviation. Middle: Resulting mass loss -ṁ as a function of time. Right: Corresponding velocity of the swimmer averaged over 10 swimmers; error bars (light blue) denote the standard deviation. The orange line is a fit U_ swim(t) = u_0exp(-t/τ_u,0)+ u_1exp(-t/τ_u,1) (see text) motivated by the existence of two swimming phases. | 2012.14364 | mdot3vexp2.jpg |
0.354834 | ac1fd165a2274ec5948d2fc7cf4157e3 | (color online) MBE layer structure of GaAs/AlGaAs single heterojunctions used in: (a) Series I, and (b) Series II. Table <ref> lists the AlGaAs barrier thickness for each wafer. The dashed red line indicates the location of the 2DEG. | 2012.14370 | Fig1.jpg |
0.443107 | dbc19f9b5d7c45a8a6630089d5d2941d | (color online) Electron mobilities for ambipolar sample L (with both n-type and p-type ohmic contacts) shown in Fig. <ref>(c), and close-up view of the multiple biased illuminations at V_topgate = -2, -4, -6, and +4 V in chronological order. The two dashed lines are otherwise identical mobility simulations, except for a difference of Δ N_sc = 8 × 10^10 /cm^2 in surface charge density, using the Boltzmann transport model described in section <ref>. The two key features to note here are: (i) all the V_topgate < 0 biased illumination mobilities fall on the same mobility curve μ(n_2d), and (ii) the mobility after the V_topgate = +4 V biased illumination increases. | 2012.14370 | Fig10.jpg |
0.432549 | 51a678cf74e041c1aa715b01bf40d4ca | (color online) Non-reversibility of transport characteristics after biased illuminations on unipolar sample M (with only n-type ohmic contacts). For all panels, the chronological order of biased illuminations and symbols used are identical to that indicated in Figure <ref>(c). (a) Unlike ambipolar Hall bars [Fig. <ref>], the mobilities after the V_topgate = +4 V biased illumination (`▴') do not increase, but remain on the same mobility curve μ(n_2d) (dashed line) as those from V_topgate < 0 biased illuminations (open symbols). (b) Likewise, the mobilities of the initial V_topgate = 0 V biased illumination (`+') are not recovered when another V_topgate = 0 V biased illumination is performed (`×') after the V_topgate < 0 biased illuminations (for clarity, only the `♢' symbols are shown), in contrast to behavior shown in Figs. <ref> and <ref> for ambipolar devices. The two dashed lines are otherwise identical μ(n_2d) mobility simulations, except for a difference of Δ N_sc = 6 × 10^10 /cm^2 in surface charge density, using the Boltzmann transport model described in section <ref>. (c) Unlike ambipolar Hall bars [see Figures <ref>(a) and <ref>(b)], the n_2d(V_topgate) curve of the original V_topgate = 0 biased illumination (`+') is not recovered after the subsequent V_topgate = 0 biased illumination (`×'). The + 0.33 V shift to a higher threshold topgate voltage to reach the same electron density is consistent with our prediction of a permanent increase in surface charge density N_sc after a V_topgate<0 biased illumination (see main text). | 2012.14370 | Fig11.jpg |
0.447913 | 9697e6bb17374974aaea4f87fb9644c5 | (color online) Effect of surface charge. (a) Experimental 2DEG mobilities at T = 1.5 K for 3 wafers (symbols) at different depths (taken from Ref. WendyPhD). Solid (red) lines are fits from the model in section III with the parameters listed in Table <ref>. The only fit parameter that changes between the three red curves is the surface charge density N_sc. All data taken in the dark. (b) Breakdown of the contributions to the mobilities shown in panel (a). The (orange) dashed line is the contribution from interface roughness (IR), common to all 3 wafers. The (green) dotted line is the contribution from the average background impurity concentration (BI), common to all 3 wafers. The (purple) solid lines are the contributions from surface charge (SC) for each wafer. For wafer A2511 (40 nm deep 2DEG), surface charge is the most significant scattering mechanism over the entire electron density range, whereas surface charge is negligible for the 310 nm deep 2DEG in wafer A2513 at all electron densities because the surface is very far away from the 2DEG. For the 80 nm deep 2DEG in wafer A2512, even though surface charge causes less scattering than either background impurities and/or interface roughness, it is still strong enough to cause the mobility to be noticeably less than wafer A2513. | 2012.14370 | Fig12.jpg |
0.447709 | e56616fefb064976a3180c2d5f618a92 | (color online) One-parameter mobility fit of 2DEGs, showcasing typical MBE chamber clean-up within a growth campaign, resulting in much-reduced background impurity concentrations. Electron mobilities, at T = 1.5 K in the dark, of 310 nm deep 2DEGs before (squares) and after (circles) the growth of some ∼ 50 wafers in the: (a) `A' chamber and (b) `V' chamber. Experimental data taken from Ref. WendyPhD. Solid (red) lines are fits from the model in section III with the parameters listed in Table <ref>. The average background impurity concentration N_bi dropped from 3.3×10^14 /cm^3 in wafer A2460 to 1.3×10^14 /cm^3 in wafer A2513 in chamber `A', and from 1.3×10^14 /cm^3 in wafer V535 to 0.7×10^14 /cm^3 in wafer V581 in chamber `V'. | 2012.14370 | Fig13.jpg |
0.478888 | 9f3fbfacb0ee42e4be03ac53bb342ccf | (color online) (a) Optical photograph of an ambipolar Hall bar with a semi-transparent 5 nm Ti topgate (in dark blue), with schematic of the circuit used for measurements. | 2012.14370 | Fig2.jpg |
0.452315 | a16e5f094d0a4c6fb29371e1e4a1f862 | (color online) Heterostructure layout for theory model. The dashed red line indicates the location of the 2DEG (at z=0), at a depth d below the wafer surface. Interface roughness irregularities are shown, along with definitions for their separation distance Λ and average height Δ. The GaAs layer is treated as semi-infinite (towards the substrate), and the 10 nm GaAs cap layer [Fig. <ref>] is treated as part of the AlGaAs barrier. | 2012.14370 | Fig3.jpg |
0.470009 | 75b71cdbc235437ca669387660de1d9d | (color online) (a) Reproducibility of mobility between two cooldowns of the same Hall bar, before illumination. (b) Reproducibility of mobility characteristics for two Hall bars from the same wafer before illumination. (c) Typical quantum Hall effect and Shubnikov-de-Haas oscillations before illumination, with visible quantized Hall plateaus at filling factors ν = 2, 3, 4, 6, and 8. | 2012.14370 | Fig4.jpg |
0.488488 | f0d5fb84dfdf4bf79bb4a7f1f13e3323 | (color online) Electron density versus topgate voltage relationships n_2d(V_topgate), before (black squares) and after (red circles) illumination, typical of Hall bars from all wafers fabricated with a: (a) polyimide insulator, and (b) SiO_2 insulator. (c) In all wafers, after the initial 5 seconds, longer illuminations did not appear to cause further change in n_2d(V_topgate). (d) After illumination, electron mobility improves. Most of the change occurs during the first 5 seconds of illumination, with eventual saturation at longer illumination times. | 2012.14370 | Fig5.jpg |
0.423307 | 8f778427d23545b78e695274a7a8ed62 | (color online) Electron mobilities before (black squares) and after (red circles) illumination for: (a)-(d) Series I wafers and (e)-(h) Series II wafers. Series I (Series II) wafers with a polyimide (SiO_2) insulator were illuminated for 4 minutes (80 seconds). Solid lines are fits to the Boltzmann transport model described in section <ref>, equation (<ref>) and equations (<ref>)-(<ref>) with fit parameter values listed in Table <ref>. Dashed lines (red or black) in panel (d) are fits to equation (<ref>), describing transport near the 2D percolation threshold. | 2012.14370 | Fig6.jpg |
0.387277 | e213c2d90e4b41088ad06a33847a84fc | Stability with time of transport characteristics in sample J, after illumination performed at time t=0 hours. Between measurements, the topgate and the 2DEG were grounded. (a) Electron density versus time, at a fixed topgate voltage. For comparison, the densities before illumination are shown (closed symbols) at t=0, for V_topgate⩽ 5 V. Error bars due to measurement uncertainty in the density (±0.03%) are smaller than the symbols and are not shown. (b) Mobility versus time, at the same fixed topgate voltages as in panel (a), with the corresponding approximate density indicated. For comparison, the mobilities before illumination are shown (closed symbols) at t=0, interpolated from the curve fit in Fig. <ref>(g) at the same corresponding electron density. Error bars due to measurement uncertainty in the mobility (±0.2%) are smaller than the symbols and are not shown. In both panels, lines are guides to the eye. | 2012.14370 | Fig7.jpg |
0.416005 | 963e63ed1bde4e7a9112749ec042dc11 | Schematic of the bandstructure of a hypothetical Hall bar from a Series I wafer [see Fig. 1] with a SiO_2 gate dielectric at V_topgate=0 (not conducting): (a) before illumination, and (b) after illumination. Although the bandstructure has some curvature, it is at least 1,000× less pronounced than in intentionally modulation-doped 2DEGs. The gate dielectric and surface charge dominate the electric field [see section IV.B], and band bending due to charged background impurities is not visible at this scale. Processes discussed in the main text are shown. | 2012.14370 | Fig8.jpg |
0.401238 | 89e230fe51fe42b1ba679b8ea7bd5e63 | (color online) Electron density versus topgate voltage after multiple biased illuminations on: (a) shallow wafer G370, and (b) deep wafer G373. Symbols in all four panels are defined in panel (c), and are in the order of biased illuminations performed during the same cooldown, with `start' being the first. In all cases, the electron density versus topgate voltage relation is shifted by the voltage at which the biased illumination was performed. Note how the characteristics of the initial `0V' biased illumination (red `+' symbols) are recovered when another biased illumination at V_topgate = 0 V is performed (black `×' symbols) after three biased illuminations are performed at V_topgate = -2 V, -4 V, and -6 V. Electron mobilities after multiple biased illuminations are shown for a: (c) shallow wafer G370, and (d) deep wafer G373. For clarity, only a selection of biased illuminations are shown in these panels. The inset in panel (d) has the same axes and units as in the main figure; it is a magnified view of the data points near n_2d = 1.1 × 10^11 /cm^2. | 2012.14370 | Fig9.jpg |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.