sentence1
stringlengths 16
446
| sentence2
stringlengths 14
436
|
|---|---|
in one-dimensional settings the cls can be classified by the integer number u of unit cells they occupy . | the cls are classified by the number u of lattice unit cells they occupy . |
to do this , we capitalize on recent advances in generative adversarial networks , which we extend to video . | to this end , we train a good paper generator using generative adversarial networks . |
this more complex structure can also qualitatively modify the intercommutation process . | furthermore , this complex microscopic structure could qualitatively modify the intercommutation process . |
from the resulting linear system we get the final conditions in the form . | we can also find the appropriate current directly from the second equation in the form . |
deep neural networks , such as cnns , have recently achieved many successes in visual recognition tasks . | convolutional neural networks have proved their dominating spot in various machine learning tasks , such as speech recognition . |
we systematically evaluate the effectiveness of these graph-based co-clustering fraud detection algorithms in identifying like farm accounts . | based on these liking patterns , we evaluated the effectiveness of existing graph based fraud detection algorithms , such as copycatch , in detecting like farm accounts . |
deep neural network approaches have attracted increased attention in image processing and computer vision , such as semantic segmentation . | convolutional neural networks have become quintessential for solving a variety of computer vision tasks such as classification . |
on the central limit theorem for stationary processes . | on mean central limit theorems for stationary sequences . |
oasis learn image similarity ranking models on top of the hand-crafted features . | oasis learn fine-grained image similarity ranking models on top of the hand-crafted features . |
because the nucleon consists of three point-like quarks , the expectation value of an opera tor with respect to the nucleon wave function eq . | the nucleon is a satisfactory laboratory to check the relation between fragmenta tion and distribution functions , since we have data both on the quark distributions of the nucleon from deep inelastic scattering . |
a completely random measure is a random measure on x whose values are independent on mutually disjoint sets . | a completely random measure is a distribution over measures on some measurable space , such that the masses g , g , . |
the nasch model is a discrete model for traffic flow . | thereby the nasch model is a minimal model in the sense that any further simplification of the model leads to an unrealistic behaviour . |
it is to be noted that sg is an infinite semigroup interval semiring . | clearly sg is an infinite non commutative non associative interval semiring . |
the mosse tracker is considered the first approach to introduce correlation filters to object tracking . | correlation filter is firstly introduced into visual tracking by the mosse tracker , in which only a single-channel feature is adopted . |
we also show an example of tip-suspended finite atomic chain to fulfill the gap between hypothetical infinite chains and experimental finite chains . | an example of optimized tip-suspended finite atomic chain is presented to bridge the gap between hypothetical infinite chains and experimental finite chains . |
deep neural networks have recently become very popular and are now successfully applied for a wide range of applications . | neural networks have attracted a significant amount of research interest in recent years due to the success of deep neural networks . |
higher order glass-transition singularities in colloidal systems with attractive interactions . | asymptotic laws near higher-order glass-transition singularities . |
dropout is a common regularization technique in neural networks to prevent overfitting and allow for a more generalizable model . | dropout is a simple and efficient technique that can be used to reduce overfitting in neural networks . |
in 1990 dhar generalized the btw model into what is now known as the abelian sandpile model . | the model was generalised by dhar in the so-called abelian sandpile model . |
convolutional neural networks have shown excellent performance in various visual recognition problems such as image classification . | recently , deep convolutional neural networks show promising performances in various computer vision tasks such as object classification , localization . |
deep neural networks perform well on many visual recognition tasks given specific training data . | convolutional neural networks have recently been shown to perform well on large scale visual recognition tasks . |
models based on deep convolutional neural networks have become the de facto standards in a wide variety of computer vision tasks , such as image classification . | recently , convolutional neural networks have been proved to be capable of dramatically boosting the performance of many mainstream computer vision problems . |
magnetite is a half metallic ferromagnet , meaning that it is an insulator in one of the . | magnetite is a well-known example of a material that undergoes a metal-insulator . |
each cnn has four convolutional layers that are followed by a batch normalization layer and a max-pooling layer . | the transition layer consists of a convolutional layer , a batch normalization layer and a relu layer sequentially . |
the study of complex networks has been one of the dominant trends in scientific research in the last decade . | the rapidly growing research on complex networks has presented a new approach to complex systems modeling and analysis . |
deep neural networks have shown impressive state-of-the-art results in the last years on numerous tasks and especially on image recognition . | convolutional neural networks have seen tremendous success across different problems including image classification . |
spatio-temporal energy models for the perception of motion . | temporal coherence theory for the detection and measurement of visual motion . |
we evaluate our proposed method on three challenging video datasets with human actions , namely hmdb51 . | we evaluate our models on three challenging action classification benchmarks , the ucf101 datasets . |
to implement this system we used the scikit-learn library . | we used the implementation from the scikit-learn package . |
we explicitly allow for covariates in the semiparametric models given below . | we explicitly allow for approximation of unknown support points . |
for example , metallic antennas , supporting localized plasmonic resonances , were shown to provide flexible solution for achieving moderate purcell enhancement and directionality in emission , eg . | for example , metallic nanoantennas supporting localized plasmonic resonances were shown to provide flexible solution for achieving moderate purcell enhancement and directionality in emission , eg . |
our proposed lung segmentation model is based on the u-net architecture . | our proposed neural network is a variant of the well known u-net architecture . |
when not shown , standard deviations are contained within symbol size . | standard deviations are contained within symbol size in all panels . |
the cosmological constant , which is the standard candidate for dark energy , can not be explained by current particle physics due to its very small value , and it is plagued with fine-tuning problems and the coincidence problem . | the cosmological constant is a parameter describing the energy density of the vacuum , and a potentially important contribution to the dy 20 time 20 time 14 namical history of the universe . |
hu et al proposes the squeeze-and-excitation block combined with attention , which can improve the representational power of a network by explicitly modeling the interdependencies between the channels of its convolutional features . | hu et al , proposed squeeze and excitation nets in which the representational power of a network is improved by explicitly modeling the interdependencies between the channels . |
game theory is the usual tool to model the interaction between selfinterested agents . | game theory is the analysis of what strategies the non-nature players will jointly choose in this situation , under different possible choice-making principles . |
also , we apply batch-normalization after every convolutional layer . | instead , we introduce a batchnormalization layer and observe a comparable performance . |
a new model calculation for high-energy gamma-ray emission from the be star outflow is introduced and the estimated gamma-ray flux considering bremsstrahlung , inverse compton scattering , and the decay of neutral pions produced in proton-proton interactions , is found to be comparable to the upper limits of these observations . | a new model for gamma-ray emission from the be star outflow has been introduced , and contributions from bremsstrahlung , inverse compton scattering , and proton-proton interactions , are calculated , with possible variations in parameters considered . |
the joint spectral radius has subsequently been found to arise in a range of mathematical contexts including control and optimisation . | the joint spectral radius has been found to arise naturally in a range of mathematical contexts including control and stability . |
a cooperative network with multiple transmitter-receiver pairs and a single eh relay is studied in by taking the spatial randomness of user locations into consideration . | a cooperative network with multiple transmitter-receiver pairs and a single energy harvesting relay is studied in by taking the spatial randomness of user locations into consideration . |
supervised deep learning has recently improved the state-of-the-art in various tasks , such as image classification . | deep learning has aided significant progress in solving various computer vision tasks such as object classification . |
quantum theory is the fundamental theory of modern physics . | for though quantum theory is a tool of unprecedented success in understanding the phenomena about us , the usual intellectual lesson derived from it is one of limitations . |
we implement the resnet-101 deep architecture as the base network for the deeplab model . | the base network for both the baseline and our method is the resnet-50 . |
the transformer architecture is based upon relying on self-attention layers to encode a sequence . | the transformer architecture relies fully on attention mechanisms , without need for recurrence or convolution . |
the electronic structure calculations and structural optimization were carried out within the density-functional formalism as implemented in the vienna ab initio simulation package . | electronic structure calculations are carried out by dft method with the generalized gradient approximation in the parametrization of perdew , burke and ernzerhof as implemented in the vienna ab-initio simulation package . |
deep neural networks have launched a profound reformation in a wide variety of fields such as image classification , detection , and segmentation . | advances in both image-based learning and language-based learning using deep neural networks have made huge strides in difficult tasks such as object recognition . |
one of the most popular is crowds , which assumes that a set of users wanting to browse the web may collaborate to submit their requests . | one of the most popular is crowds , which contemplates that a group of users wanting to browse the web will collaborate to submit their requests . |
recently , neural networks have achieved very impressive success on a wide range of fields like computer vision . | deep learning has been very successful in many applications such as computer vision . |
lu et al propose to learn language priors from semantic word embedding to fine-tune visual relationships . | to further exploit robust relationship prediction with limited training samples , lu et al proposed a visual relationship model with language priors . |
the crosscorrelation matrix c of price returns is analyzed to investigate the interactions between stocks 12 , 18 , . | the spacial structure of the stock markets is explored by investigating the cross-correlation of stocks 12 , 18 , . |
one can efficiently compute σ-basis using the algorithm pm-basis of . | if all matrices involved are of degree d , then we use the σ-basis algorithm of . |
network features are extracted by constructing specific social networks , such as diffusion networks or co-occurrence networks . | network features are extracted by constructing specific social networks , such as diffusion networks . |
recently , bergshoeff , hohm and townsend proposed a parity conserving theory of gravity in three dimensions , which is defined by adding certain curvature-squared terms to the einstein-hilbert action . | recently , bergshoeff , hohm and townsend constructed a particular 3d gravity theory with a massive spin-2 propagating mode , which is often called new massive gravity theory . |
moreover , many recent works showed that neural networks can be successfully used in a number of tasks in natural language processing , such as machine translation . | recently , the recurrent neural networks has been widely used in various sequence prediction tasks , including natural language generation . |
the quantum number m is a conserved quantity obeying sz . | this quantum number is the topological charge , which is temperature 6 independent . |
dthis is the eastern of the two central cds . | dthis is the estimated photon index based on the x-ray band ratio . |
in baseline-1 , features were extracted from the output of the last convolutional layer of the vgg-16 network trained on imagenet . | in baseline-1 , features were extracted from the last convolutional layer of the vgg-16 network trained on imagenet . |
also , recent deep approaches in person re-identification are found to unify feature learning and metric learning . | recently , deep learning methods have been applied to person re-identification . |
liu et al used two cascaded cnns and trained support vector machines to separate the processes of face localization and attribute prediction . | liu et al combined three networks to first localize faces with two localization networks and then extract features using the attribute network . |
an asterisk denotes data which are not known . | the asterisk is the value discussed in detail in the text . |
mapping cones and exact sequences in kk-theory . | equivariant kk-theory and the novikov conjecture . |
saglia rp , kronawitter a , gerhard o , bender r . | matsushita k , belsole e , finoguenov a , bohringer h . |
oeckl , twisting of quantum differentials and the planck scale hopf algebra , commun . | witten , non-abelian bosonization in two dimensions , commun . |
in order to make more clear the continuum hypothesis , the study of the microto-continuum limit for first and second order models has been proposed in respectively . | in order to justify and make more clear the continuum hypothesis , the study of the discrete-to-continuum limit for second order models has been proposed in . |
dispersion of recovered properties of artificial clusters , comparing various wide and medium-band hst filters , as indicated in the legend . | dispersion of recovered properties of artificial clusters , assuming availability of ubih magnitudes and varying observational errors , as indicated in the legend . |
hyers gave a first affirmative partial answer to the question of ulam for banach spaces . | hyers gave a positive answer to the question of ulam for banach spaces . |
recently , deep neural networks have achieved impressive results for many image classification tasks . | onvolutional neural networks have achieved remarkable performance on vision problems such as image classification . |
this leads to a chiral symmetry , which is the rotation of fermion fields by eiγ5θ . | we shall assume that the chiral symmetry is a good symmetry in our analysis . |
two of the most popular deep generative models are the variational autoencoder and the generative adversarial network . | deep generative models come in two broad flavours -variational autoencoders , and generative adversarial networks . |
the families appearing are constructed as families of algebras of krichever-novikov type . | in special cases the witt and virasoro algebra appear as degenerations of the krichever-novikov algebras . |
if a dynamical system maintains a partial ordering of states along trajectories of the system , it is said to be monotone . | a dynamical system is monotone if it maintains a partial ordering of states along trajectories of the system . |
euclidean structures of simplicial surfaces and hyperbolic volume . | discrete riemann surfaces and the ising model . |
the dark energy is a true cosmological constant , strictly unchanging throughout space and time . | dark energy is a cosmological constant as a stationary value of a potential of self-interaction of ψ field . |
our construction is motivated by the work in , which studied inner approximations of the cone of positive semidefinite matrices based on the cones of diagonally dominant and scaled diagonally dominant matrices . | our approach is motivated by the recent work in that uses the cone of scaled diagonally dominant matrices for innerapproximating the cone of positive semidefinite matrices . |
second , on-chip sram and register-file accesses dominate the energy consumption when accelerating dnns . | second , energy consumption for dnn acceleration is usually dominated by data accesses to on-chip storage and off-chip memory . |
these polynomials are the stieltjes-wigert polynomials in the hermitian case . | the orthogonal polynomials of this matrix ensemble are the stieltjes-wigert polynomials . |
wikispeedia 6 is an online crowd sourcing game designed to measure semantic distances between 2 wikipedia pages using the paths taken by humans to reach from one page to the other . | wikispeedia wikispeedia 5 is an online crowd sourcing game designed to measure semantic distances between two wikipedia pages using the paths taken by humans to reach from one page to the other . |
reinforcement learning is a type of machine learning in which an agent learns from its experience in an environment to maximize some cumulative reward . | reinforcement learning is an area of machine learning that enables artificial agents to identify optimal behavioral policies through interactions with their environment . |
all the parameters of the decoder network are initialized with xavier . | the parameters in last fully-connected layer are initialize with xavier initialization . |
cosmic strings are topological defects that could be produced at phase transitions in the early universe , . | cosmic strings are line-like topological defects which may form during phase transitions in the early universe . |
in the last few years , generative adversarial networks , initially proposed in , gained much attention in the image processing domain . | recently , generative adversarial network has drawn a lot of attention in image generation . |
the vertical solid and dashed grey lines indicate gluino and squark masses respectively . | the horizontal and curved grey lines indicate gluino and squark mass contours , respectively . |
in deeplab series , an atrous spatial pyramid pooling module is introduced which employs multiple parallel filters with different dilation rates to collect multi-scale information . | the atrous spatial pyramid pooling is used in deeplabv2 to capture multi-scale information via equipping parallel atrous convolutional layers . |
recent work , however , has demonstrated that dnns are vulnerable to adversarial perturbations . | however , recently it has been shown that dnns are susceptible to attacks by adversarial examples . |
convolutional neural networks have achieved tremendous progress on many pattern recognition tasks , especially large-scale images recognition problems . | in recent years , convolutional neural networks have become the de facto standard in many computer vision tasks , such as image classification and object detection . |
variable-speed-of-light cosmological models were proposed for solving the initial value problems in the standard big bang model . | variable-speed-of-light cosmological models were proposed for solving the cosmological problems of the standard big bang model . |
convolutional neural networks are the state of the art in image classification problems . | recently , deep neural networks have demonstrated impressive results in image classification . |
deep neural networks are powerful learning models which have been successfully applied to vision , speech and many other tasks . | in recent years , deep convolutional neural networks have proven to be highly effective general models for a multitude of computer vision problems . |
from the figures , we can see that there is a shift from metal-rich stars to metal-poor ones with the increasing of apparent magnitude . | the figures show that there is a stochastic layer near separatrices , which at first increases , and then decreases together with the area of possible motion . |
the electron-core interactions are described by the projector augmented wave method , while we use perdew-burke-ernzerhof parametrization of the generalized gradient approximation for the xc functional . | the projector augmented wave method is used to calculate the wave functions , and the generalized gradient approximation formulated by perdew , burke and ernzerhof is taken into account for the exchange-correlation functional . |
for the feature extractor component , we initialize it by the pretrained alexnet model . | for the svm ranker and binary dominance , we generate cnn features from the fc7 layer of alexnet . |
recent advances in machine learning and data analytics have yielded transformative results across diverse scientific disciplines , including image recognition . | recent advances in deep learning have revolutionized the application of machine learning in areas such as computer vision , speech recognition and natural language processing . |
this completion is the first step in a bootstrapping process of understanding the full completion . | this completion is the goal of the lines below . |
the neutralino is a weakly interacting massive particle . | this is because the neutralino is a majorana spinor , and therefore can decay equally into leptons and antileptons . |
notice that the least squares formulations in and are exactly analogous to the optimal homologous chain problem from . | notice that the least squares formulations in and are exactly the optimal homologous chain problem from . |
the casimir force is the dominant force between two neutral non-magnetic objects in the range of interest so that any new force would appear as a difference between experimental measurements and theoretical expectations of the casimir force . | more generally , the casimir force is a result of the boundedness or deviation from a euclidean topology of or in the quantum vaccuum . |
all networks are trained for 40 epochs using the adam optimizer with standard parameters . | these networks are trained with adam optimizer for 200 epochs . |
in addition , bampis et al studied a heterogeneous multiprocessor preemptive problem , in which it was assumed that each processor had a different speed-to-power function . | bampis et al studied the heterogeneous multiprocessor preemptive problem where every processor i has a different speed-to power function , s α , and both the life interval and the work of jobs are processor dependent . |
in recent years , convolutional neural networks have become the dominant approach for a variety of computer vision tasks , eg , image classification . | deep learning based models have emerged as an extremely powerful framework to deal with different kinds of vision problems including image classification . |
remember that strain is the dominant cause of orientational freezing and that the strain energy is the same whether the polarization points up or down . | when the strain is larger than one , there is a competition between coarsening and shear-induced deformations of domains . |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.