sentence1
stringlengths 16
446
| sentence2
stringlengths 14
436
|
|---|---|
deep convolutional neural networks have achieved great success in computer vision related tasks such as image classification , and so on . | convolutional neural networks have recently provided state of the art results for several recognition tasks including object recognition . |
since string theory is a proposed fundamental theory of quantum gravity , one may expect that stringy corrections to the low energy effective theory resolve causality violations . | string theory is a unifying principle that blends these disparate n notions together . |
when applied to the binary linear codes , the three generalized bounds are reduced to the conventional ones . | when applied to binary linear codes , the triangle distance spectrum is reduced to the conventional weight distribution . |
this is reminiscent of the more recent generative adversarial networks , in which a generator relies on a discriminator to improve . | this can possibly be overcome with the use of generative adversarial network where the loss is derived from a discriminator network . |
we pair each cityscpaes image with its nearest neighbor in gta , retrieved by the inception-resnet-v2 features . | for imagenet , we use the top 800 pca components of some convolutional features extracted from inception-resnet-v2 . |
hu , crystal bases and simple modules for hecke algebra of type dn . | geck , representations of hecke algebras at roots of unity . |
moreover , we numerically determine the full nonperturbative effects , which give further evidence of supersymmetry breaking in the double scaling limit . | furthermore , we numerically determine the full nonperturbative effects by recursive evaluation of orthogonal polynomials . |
recently , deep convolutional neural networks show promising performances in various computer vision tasks such as object classification , localization . | convolutional neural networks have witnessed great improvement on a series of vision tasks such as object classification . |
convolutional neural networks are enabling major advancements in a range of machine learning problems . | since 2012 , neural networks and deep architectures have proven very effective in application areas such as computer vision . |
if that starting configuration consists of just closed paths , a plaquette update algorithm will subsequently also generate only loop configurations . | since the starting configuration is a purely gravitational background , the type iia configuration involves a ten dimensional metric , dilaton and a rr 1-form potential , all other fields vanishing . |
keskar et al suggested that there is a generalization problem for large-batch training . | it is also shown in keskar et al that training accuracies for both large and small batch methods are comparably good . |
the proof uses reduction from the vertex cover problem , shown to be np -hard by karp in . | the proof uses a reduction from the subset sum problem , shown to be np-hard by karp in . |
references examined the multiple access channel with confidential messages where two transmitters try to keep their messages secret from each other while communicating with a common receiver . | references examined the multiple access channel with confidential messages , where two transmitters try to keep their messages secret from each other while communicating with a common receiver . |
we prove that our bounds are tighter than those obtained by audenaert , mosonyi and verstraete in , for a wide range of threshold values of the type i error , which is of practical relevance . | we prove that our bounds are tighter than those obtained by audenaert , mosonyi and verstrate in , for a wide range of threshold values of the type i error which is of practical relevance . |
single photon sources are a fundamental component of the toolbox for quantum information technologies that promise transformational advances in the communication and processing of information . | entangled photons are essential for many fundamental quantum optics experiments , as well as a key resource in quantum communication . |
between each convolutional layer and following non-linearity we use batch normalization . | we apply batch-normalization after each convolution and use leaky rectified linear units as activations . |
recently , deep convolutional neural networks show promising performances in various computer vision tasks such as object classification , localization . | however , deep convolutional neural networks have demonstrated impressive performances on different tasks such as object recognition . |
we apply this theory to the class of uneven margin losses , and characterize when these losses are properly calibrated . | we apply this framework to derive surrogate regret bounds for cost-sensitive classification and arbitrary losses . |
in recent years , convolutional neural networks have shown excellent performance on classification problems when large-scale labeled datasets are available . | following the prevalence and advances of cmos active pixel sensing , deep convolutional neural networks have already achieved good performance in aps-based object classification problems . |
this was first demonstrated by werner , who presented a class of entangled states which admit a lhv model for arbitrary projective measurements . | this was first shown by werner , who presented a class of entangled states whose statistics can be reproduced by a local hidden-variable model , considering arbitrary projective measurements . |
neural networks have been widely adopted in many scenarios , achieving state-of-the-art results in numerous tasks . | deep neural networks have achieved significant improvements in many computer vision tasks . |
deep convolutional neural networks have advanced to show the state-of-the-art performance in image processing and computer vision applications . | deep neural networks have made great strides in many computer vision tasks such as image classification . |
thus , the lfs of the rest-frame spire bands show qualitatively similar evolution as the lfs of the pacs bands . | the redshift evolution of the rest-frame spire band lfs is similar to that of the pacs bands . |
models based on convolutional neural networks and recurrent neural networks have achieved remarkable performance in many tasks , such as image classification . | convolutional neural networks have shown significant success in challenging tasks in image classification and recognition . |
hed output edge predictions from intermediates layers , which are deeply supervised , and fuses the predictions by linear weighting . | hed output edge predictions from intermediate layers , which are deeply supervised , and fuses the predictions by linear weighting . |
convolutional neural networks have been an active research topic in computer vision mostly because they have achieved state-of-the-art results in numerous tasks . | convolutional neural networks are a well-established approach in computer vision contributing to the success of applications in image segmentation , object detection , and image classification . |
the corresponding sigma models are torsionless but their metric is not flat . | the corresponding sigma models are torsionless , their metrics are flat . |
now let us define the local coordinates of various patches in terms of homogeneous coordinates φi . | we define the local coordinates of various patches in terms of homogeneous coordinates φi . |
the proof of the inequality for a connected horizon follows the inverse mean curvature flow argument of huisken-ilmanen . | this follows from the inverse mean curvature flow argument of huisken and ilmanen in the proof of the riemannian positive mass theorem . |
recent work has focused on the pmu data utilization at the transmission level to improve the wide-area monitoring , protection , and control . | there has been a lot of work focused on using pmu data at the transmission level to improve wide-area monitoring , protection and control . |
it had been shown that long , perfectly matching dsrnas are extensively edited upon transfection to mammalian cells . | indeed , such long and perfectly matching dsrnas are extensively edited upon transfection to mammalian cells in vitro . |
guerrero , walter wilcox , and joe christensen . | lee , scott moerschbacher , and walter wilcox . |
the resulting amplitude satisfies gauge invariance . | obviously the amplitude of hm satisfies the gauge invariance . |
in , the authors learn projections between low-dimensional subspaces of 3d patch-volumes of low-and high-resolution , using ridge regression . | in , the authors show that a 3d patch-volume resides on a low-dimensional subspace and propose to learn a projection between low-and high-resolution subspaces of patch-volumes using ridge-regression . |
now we proceed on to define smarandache parallel . | now we proceed onto define smarandache analogue . |
a variety of papers have considered modeling spatial relationships in natural images . | a variety of papers have considered modeling spatial relations in natural images . |
convolutional neural networks have made great progress in various fields , such as object classification , detection and character recognition . | in recent years , deep neural networks have been applied to many areas and have achieved huge success in different domains such as image classification . |
at lower redshifts , there is a larger difference , with the distorted galaxies revealing a higher star formation rate . | at higher redshifts there is a suggestion that sub-mm galaxies are signficantly fainter at k than their radio-selected counterparts , but at present it is unclear whether this indicates a significant difference in stellar mass or the increasing impact of dust obscuration on the rest-frame light from the sub-mm hosts . |
the generalized-gradient approximation to the exchange-correlation functional proposed by perdew , burke and ernzerhof was used . | the generalized gradient approximation of perdew-burke-ernzerhof was adopted for the exchange-correlation functional . |
we use resnet-50 as the backbone network and feature pyramid network as the detection framework . | we adopt a residual neural network with 50 layers resnet50 as the base architecture for smileynet . |
deep learning methods have achieved impressive performance in object recognition and classification by using large networks trained with millions of data examples . | convolutional neural networks have achieved great success in many fields , such as object classification , face recognition . |
cosmic strings are thin filaments of topologically-trapped higgs field energy which may have formed at a symmetry-breaking phase transition in the early universe . | cosmic strings are line-like topological defects which may form during phase transitions in the early universe . |
the algorithm uses an alternating method developed in to replace the twosided equation by two opposite inequalities and then alternately solve them to achieve more and more accurate estimates for a solution to the equation . | the algorithm uses an alternating method developed by to replace the equality constraint with two opposite inequalities , which are solved iteratively to provide progressively better bounds for a solution . |
rle , a simple compression method was first used for coding pictures . | run-length encoding , a simple compression method was first used for coding pictures . |
it was shown by shannon that the presence of noiseless feedback does not increase the capacity of point-to-point memoryless channels . | it is well known that feedback can not increase the capacity of memoryless point-to-point channels . |
studies in recent years employing deep neural networks has high performance in the classification task , which has lead to significant progress in a variety of real-world applications such as image classification . | in recent years , deep learning methods , such as convolutional neural networks and recurrent neural networks , and their variations , have demonstrated excellent performance in visual and multi-label classification . |
the deformable part model from felzenszwalb et al is a representative approach for object detection . | a latent svm model is used by felszenzwalb et al to detect objects using deformable part models . |
the perdew-burke-ernzerhof exchange-correlation functional in the generalized gradient approximation was employed . | the perdew-burke-ernzerhof exchange correlation functional and a double-ζ basis including polarization functions were employed . |
on the counting complexity of propositional circumscription . | on the intertranslatability of non-monotonic logics . |
his research interests are widely in wireless communications , digital signal processing and convex optimization and its applications . | his research interests include probability theory , optimization , information theory and wireless communication . |
the benefit is the fact that , for triploids , mutations that happen in loci where the harmful allele is not dominant need to appear in all three bit-strings to become active . | another benefit is the fact that network packets now have the unprecedented ability to control their own processing . |
recent development in deep learning has extended the convolutional networks idea from image to graphs . | recently proposed graph convolutional networks provide a framework to apply deep neural networks to graph-structured data . |
many linguists have formulated various constraints to define a general rule for code-switching . | many linguists formulated various constraints to define a general rule for code-switching . |
this nonlocality is a natural consequence of the presence of the uv divergence . | technically , non-locality is a direct consequence of the non-triviality of elko spin sums . |
here we study the key question of whether excitation transport on light-harvesting complexes shows quantum speedup . | here we compare the dynamics in these light harvesting systems to the dynamics of quantum walks , in order to elucidate the limits of such quantum speedups . |
recently , the sampling recovery for sobolev and besov classes having mixed smoothness has been investigated in . | recently , the sampling recovery for periodic besov classes having mixed smoothness has been investigated in . |
in , sierra studied gr -a 1 , the category of finitely generated z-graded right a 1 -modules . | in , sierra studied gr -a 1 , the category of finitely generated graded right a 1 -modules . |
for details of the underlying numerical method , we refer the reader to . | for a more thorough discussion on the branch-cut and its computation , we refer readers to . |
also , presented a generalization of hausdorff distance , gromov-hausdorff distance , and the space of metric spaces in the form of a categorical view . | also presented a generalization of hausdorff distance , gromov-hausdorff distance , and the space of metric spaces in the form of categorical view . |
more recently , deep learning approaches have grown out of the bio-inspired roots of anns to become the state-of-the-art machine learning method . | in recent years , deep convolutional neural networks have set the state-of-the-art on a broad range of computer vision tasks . |
in recent years , the accuracy of object detection has been dramatically improved thanks to the advance of deep convolutional neural network . | the performance of visual scene recognition tasks has been significantly boosted by recent advances of deep learning algorithms . |
these projects are from the travistorrent dataset and have at least 50 commits in at least 1 java file . | these projects also appear in travistorrent and have at least 50 commits in java files . |
we implement our model using the torch7 framework . | implementation we implement our model under the torch7 framework . |
the choice of the parametric model p θ should account for the standard trade-off between expressivity of the model and overfitting . | the choice of the parametric model pshould account for the standard trade-off between capacity of the model and overfitting . |
characterize those lie algebras which are s-strongly lie simple . | characterize those commutative groupoid rings which do not have n-capacitor . |
the literature indicates that the ccmbased algorithms have a superior performance to algorithms based on the constrained minimum variance criterion . | the designs based on the constrained constant modulus criterion have shown increased robustness against signature mismatch and improved performance over constrained minimum variance approaches . |
the first is to learn important features and then apply classification to infer pixel labels . | the first is to learn image features and apply pixel classification . |
when energy and information are exchanged between two systems , the dynamics of energy exchange does not uniquely determine the information exchanged . | so when energy is exchanged between two systems , information is also exchanged , but the dynamics of energy exchange does not uniquely determine the information exchanged . |
the computed trajectories do well converge to the results of the effective feedback master equation . | the average of the trajectories converges to the effective feedback master equation result . |
adversarial examples for neural networks are small perturbations of inputs carefully crafted to fool trained models . | adversarial examples have recently been considered for neural networks , where the input is typically perturbed locally in order to find counterexamples . |
since the heat kernel is a positive distribution , it is a measure and follows from the fact that the heat kernel is a distributional solution of the corresponding degenerate heat equation . | physically , the heat kernel is a crucial element of the spectral action approach . |
recall that finite groups generated by complex reflections have been classified by shephard and todd . | finite linear groups generated by complex reflections were classified by shephard and todd . |
many exotic optical properties have been demonstrated experimentally based on various metamaterials , including negative refractive index . | numerous diffraction elements have been recently proposed for controlling light refraction and for generation of unconventional optical beams . |
deep convolutional neural networks have shown tremendous success in a variety of computer vision tasks , such as image classification . | deep neural networks have been very successful in large-scale recognition and classification tasks , some even surpassing human-level accuracy . |
shokri et al propose the first membership inference attack against machine learning models . | shokri et al are among the first to perform effective membership inference against ml models . |
deep neural networks , in particular convolutional neural networks , have been used with great success for perceptual tasks such as image classification . | specifically , convolutional neural networks have shown their powerful abilities on image representation . |
massive mimo is the key technology for increasing the se in future cellular networks , by virtue of beamforming and spatial multiplexing . | in addition , the large antenna gains used for beamforming makes massive mimo one of the most promising methods for increasing the spectral efficiency of future cellular networks . |
additionally , we compile the model using the adamax optimizer . | additionally , we use adam as our optimization method and batch-normalization layers . |
in recent years , deep neural networks have demonstrated impressive performance improvements on a wide range of challenging machine learning tasks . | deeper convolutional neural networks have obtained state-of-the-art results in many image classification tasks . |
a monopole is a highly coherent state of many gauge quanta and emission of a monopole by a black hole will be highly suppressed . | the homotopy class m is called the degree of the monopole . |
the wiretap channel was first introduced by wyner , in which a legitimate transmitter wishes to send messages to a legitimate receiver secretly in the presence of an eavesdropper . | the wiretap channel was introduced by wyner as a discrete memoryless broadcast channel where the sender , alice , transmits confidential messages to a legitimate receiver bob , in the presence of an eavesdropper eve . |
imaging and self-calibration were performed using the difmap software package . | cleaning , phase and amplitude self-calibration , and hybrid imaging were performed in the caltech difmap package . |
among these , silicon is the best measured one in the icm . | silicon is a promising candidate for constructing a cavity-phoniton system . |
berry was supported by the arc centre for quantum computer technology , the department of physics at macquarie university , and the institute for quantum computing . | wiseman was supported by the arc centre for quantum computer technology , and the centre for quantum dynamics at griffith university . |
the asymmetry of the pulses gives indeed a measure of this type . | asymmetry is the twist-3 fragmentation function of the pion . |
in recent years , neural network approaches have significantly advanced the state of the art in computer vision tasks such as classification . | in the past decade , the advent of large-scale datasets and improvements in training deep neural networks have enabled massive advances in computer vision , especially in image classification . |
evolutionary multi-objective optimization algorithms have proven their benefits for solving multi-objective problems . | evolutionary multi-objective optimization algorithms have demonstrated their ability for solving multi-objective optimization problems . |
stewart et al use a recurrent framework to sequentially detect people . | stewart et al employed a recurrent lstm layer for people detection . |
for some hilbert space h is a completely contractive banach algebra . | the hilbert space h is the cauchy completion of v with this inner product . |
because quantum field theory is a kind of relativistic theory , it should obey some fundamental principles of the special theory of relativity , such as principle of special relativity and principle of invariance of light speed . | in quantum field theory there is a standard procedure for integrating out high energy degrees of freedom and obtaining an effective theory at low energy . |
orpl brings opportunistic routing into rpl and improves the performance of rpl . | similarly , orpl incorporates opportunistic routing with rpl to achieve low latency , robustness , and good scalability . |
the history is a bit-string of length m recording the minority option for the most recent m time steps . | each history is a set of possible measurement outcomes , for observables at different times . |
chunk scheduling problems in uncoded peer-topeer networks , as opposed to point-to-multipoints , are considered in . | chunk scheduling problems in uncoded peer-to-peer networks , as opposed to pmps , are considered in . |
clean power spectra of ekar observations after correction for the long-term modulation . | clean power spectra of wiro data after correction for the long-term trends . |
the pascal voc 2007 dataset contains 9,963 images from 20 object classes . | the pascal voc dataset contains box annotations over 20 object categories . |
in recent years , convolutional neural networks have achieved outstanding performance in a variety of machine learning tasks , especially in computer vision , such as image classification . | designing deeper and wider convolutional neural networks has led to significant breakthroughs in many machine learning tasks , such as image classification . |
a gating network will be parity preserving if its individual gates are parity preserving . | and a gating network will be parity preserving if its individual gate is parity preserving . |
reinforcement learning is the problem of learning from interaction with the environment to achieve a goal . | reinforcement learning is a type of machine learning which learns through interaction with the environment . |
inflation is the leading paradigm to solve the puzzles in the hot big bang model . | inflation is a successful theory to solve some of the problems of the otherwise successful big-bang model . |
in this section , we evaluate the proposed approach on the challenging kitti object detection benchmark dedicated to autonomous driving . | in this section , we present an evaluation of the robustness of popular computer vision algorithms on kitti to demonstrate the usefulness of our rain rendering methodology . |
Subsets and Splits