May 5, 2019

2872 words 14 mins read

Paper Group ANR 555

Visual Fashion-Product Search at SK Planet. Nonnegative autoencoder with simplified random neural network. Fast Learning with Nonconvex L1-2 Regularization. Probabilistic Reasoning via Deep Learning: Neural Association Models. Enablers and Inhibitors in Causal Justifications of Logic Programs. Deep Semi-Supervised Learning with Linguistically Motiv …

Visual Fashion-Product Search at SK Planet


Title	Visual Fashion-Product Search at SK Planet
Authors	Taewan Kim, Seyeong Kim, Sangil Na, Hayoon Kim, Moonki Kim, Byoung-Ki Jeon
Abstract	We build a large-scale visual search system which finds similar product images given a fashion item. Defining similarity among arbitrary fashion-products is still remains a challenging problem, even there is no exact ground-truth. To resolve this problem, we define more than 90 fashion-related attributes, and combination of these attributes can represent thousands of unique fashion-styles. The fashion-attributes are one of the ingredients to define semantic similarity among fashion-product images. To build our system at scale, these fashion-attributes are again used to build an inverted indexing scheme. In addition to these fashion-attributes for semantic similarity, we extract colour and appearance features in a region-of-interest (ROI) of a fashion item for visual similarity. By sharing our approach, we expect active discussion on that how to apply current computer vision research into the e-commerce industry.
Tasks	Semantic Similarity, Semantic Textual Similarity
Published	2016-09-26
URL	http://arxiv.org/abs/1609.07859v6
PDF	http://arxiv.org/pdf/1609.07859v6.pdf
PWC	https://paperswithcode.com/paper/visual-fashion-product-search-at-sk-planet
Repo
Framework

Nonnegative autoencoder with simplified random neural network


Title	Nonnegative autoencoder with simplified random neural network
Authors	Yonghua Yin, Erol Gelenbe
Abstract	This paper proposes new nonnegative (shallow and multi-layer) autoencoders by combining the spiking Random Neural Network (RNN) model, the network architecture typical used in deep-learning area and the training technique inspired from nonnegative matrix factorization (NMF). The shallow autoencoder is a simplified RNN model, which is then stacked into a multi-layer architecture. The learning algorithm is based on the weight update rules in NMF, subject to the nonnegative probability constraints of the RNN. The autoencoders equipped with this learning algorithm are tested on typical image datasets including the MNIST, Yale face and CIFAR-10 datasets, and also using 16 real-world datasets from different areas. The results obtained through these tests yield the desired high learning and recognition accuracy. Also, numerical simulations of the stochastic spiking behavior of this RNN auto encoder, show that it can be implemented in a highly-distributed manner.
Tasks
Published	2016-09-25
URL	http://arxiv.org/abs/1609.08151v2
PDF	http://arxiv.org/pdf/1609.08151v2.pdf
PWC	https://paperswithcode.com/paper/nonnegative-autoencoder-with-simplified
Repo
Framework

Fast Learning with Nonconvex L1-2 Regularization


Title	Fast Learning with Nonconvex L1-2 Regularization
Authors	Quanming Yao, James T. Kwok, Xiawei Guo
Abstract	Convex regularizers are often used for sparse learning. They are easy to optimize, but can lead to inferior prediction performance. The difference of $\ell_1$ and $\ell_2$ ($\ell_{1-2}$) regularizer has been recently proposed as a nonconvex regularizer. It yields better recovery than both $\ell_0$ and $\ell_1$ regularizers on compressed sensing. However, how to efficiently optimize its learning problem is still challenging. The main difficulty is that both the $\ell_1$ and $\ell_2$ norms in $\ell_{1-2}$ are not differentiable, and existing optimization algorithms cannot be applied. In this paper, we show that a closed-form solution can be derived for the proximal step associated with this regularizer. We further extend the result for low-rank matrix learning and the total variation model. Experiments on both synthetic and real data sets show that the resultant accelerated proximal gradient algorithm is more efficient than other noncovex optimization algorithms.
Tasks	Sparse Learning
Published	2016-10-29
URL	http://arxiv.org/abs/1610.09461v3
PDF	http://arxiv.org/pdf/1610.09461v3.pdf
PWC	https://paperswithcode.com/paper/fast-learning-with-nonconvex-l1-2
Repo
Framework

Probabilistic Reasoning via Deep Learning: Neural Association Models


Title	Probabilistic Reasoning via Deep Learning: Neural Association Models
Authors	Quan Liu, Hui Jiang, Andrew Evdokimov, Zhen-Hua Ling, Xiaodan Zhu, Si Wei, Yu Hu
Abstract	In this paper, we propose a new deep learning approach, called neural association model (NAM), for probabilistic reasoning in artificial intelligence. We propose to use neural networks to model association between any two events in a domain. Neural networks take one event as input and compute a conditional probability of the other event to model how likely these two events are to be associated. The actual meaning of the conditional probabilities varies between applications and depends on how the models are trained. In this work, as two case studies, we have investigated two NAM structures, namely deep neural networks (DNN) and relation-modulated neural nets (RMNN), on several probabilistic reasoning tasks in AI, including recognizing textual entailment, triple classification in multi-relational knowledge bases and commonsense reasoning. Experimental results on several popular datasets derived from WordNet, FreeBase and ConceptNet have all demonstrated that both DNNs and RMNNs perform equally well and they can significantly outperform the conventional methods available for these reasoning tasks. Moreover, compared with DNNs, RMNNs are superior in knowledge transfer, where a pre-trained model can be quickly extended to an unseen relation after observing only a few training samples. To further prove the effectiveness of the proposed models, in this work, we have applied NAMs to solving challenging Winograd Schema (WS) problems. Experiments conducted on a set of WS problems prove that the proposed models have the potential for commonsense reasoning.
Tasks	Natural Language Inference, Transfer Learning
Published	2016-03-24
URL	http://arxiv.org/abs/1603.07704v2
PDF	http://arxiv.org/pdf/1603.07704v2.pdf
PWC	https://paperswithcode.com/paper/probabilistic-reasoning-via-deep-learning
Repo
Framework

Enablers and Inhibitors in Causal Justifications of Logic Programs


Title	Enablers and Inhibitors in Causal Justifications of Logic Programs
Authors	Pedro Cabalar, Jorge Fandinno
Abstract	To appear in Theory and Practice of Logic Programming (TPLP). In this paper we propose an extension of logic programming (LP) where each default literal derived from the well-founded model is associated to a justification represented as an algebraic expression. This expression contains both causal explanations (in the form of proof graphs built with rule labels) and terms under the scope of negation that stand for conditions that enable or disable the application of causal rules. Using some examples, we discuss how these new conditions, we respectively call “enablers” and “inhibitors”, are intimately related to default negation and have an essentially different nature from regular cause-effect relations. The most important result is a formal comparison to the recent algebraic approaches for justifications in LP: “Why-not Provenance” (WnP) and “Causal Graphs” (CG). We show that the current approach extends both WnP and CG justifications under the Well-Founded Semantics and, as a byproduct, we also establish a formal relation between these two approaches.
Tasks
Published	2016-02-22
URL	http://arxiv.org/abs/1602.06897v1
PDF	http://arxiv.org/pdf/1602.06897v1.pdf
PWC	https://paperswithcode.com/paper/enablers-and-inhibitors-in-causal
Repo
Framework

Deep Semi-Supervised Learning with Linguistically Motivated Sequence Labeling Task Hierarchies


Title	Deep Semi-Supervised Learning with Linguistically Motivated Sequence Labeling Task Hierarchies
Authors	Jonathan Godwin, Pontus Stenetorp, Sebastian Riedel
Abstract	In this paper we present a novel Neural Network algorithm for conducting semi-supervised learning for sequence labeling tasks arranged in a linguistically motivated hierarchy. This relationship is exploited to regularise the representations of supervised tasks by backpropagating the error of the unsupervised task through the supervised tasks. We introduce a neural network where lower layers are supervised by junior downstream tasks and the final layer task is an auxiliary unsupervised task. The architecture shows improvements of up to two percentage points F1 for Chunking compared to a plausible baseline.
Tasks	Chunking
Published	2016-12-29
URL	http://arxiv.org/abs/1612.09113v1
PDF	http://arxiv.org/pdf/1612.09113v1.pdf
PWC	https://paperswithcode.com/paper/deep-semi-supervised-learning-with
Repo
Framework

SSSC-AM: A Unified Framework for Video Co-Segmentation by Structured Sparse Subspace Clustering with Appearance and Motion Features


Title	SSSC-AM: A Unified Framework for Video Co-Segmentation by Structured Sparse Subspace Clustering with Appearance and Motion Features
Authors	Junlin Yao, Frank Nielsen
Abstract	Video co-segmentation refers to the task of jointly segmenting common objects appearing in a given group of videos. In practice, high-dimensional data such as videos can be conceptually thought as being drawn from a union of subspaces corresponding to categories rather than from a smooth manifold. Therefore, segmenting data into respective subspaces — subspace clustering — finds widespread applications in computer vision, including co-segmentation. State-of-the-art methods via subspace clustering seek to solve the problem in two steps: First, an affinity matrix is built from data, with appearance features or motion patterns. Second, the data are segmented by applying spectral clustering to the affinity matrix. However, this process is insufficient to obtain an optimal solution since it does not take into account the {\em interdependence} of the affinity matrix with the segmentation. In this work, we present a novel unified video co-segmentation framework inspired by the recent Structured Sparse Subspace Clustering ($\mathrm{S^{3}C}$) based on the {\em self-expressiveness} model. Our method yields more consistent segmentation results. In order to improve the detectability of motion features with missing trajectories due to occlusion or tracked points moving out of frames, we add an extra-dimensional signature to the motion trajectories. Moreover, we reformulate the $\mathrm{S^{3}C}$ algorithm by adding the affine subspace constraint in order to make it more suitable to segment rigid motions lying in affine subspaces of dimension at most $3$. Our experiments on MOViCS dataset show that our framework achieves the highest overall performance among baseline algorithms and demonstrate its robustness to heavy noise.
Tasks
Published	2016-03-14
URL	http://arxiv.org/abs/1603.04139v2
PDF	http://arxiv.org/pdf/1603.04139v2.pdf
PWC	https://paperswithcode.com/paper/sssc-am-a-unified-framework-for-video-co
Repo
Framework

Visual Tracking via Reliable Memories


Title	Visual Tracking via Reliable Memories
Authors	Shu Wang, Shaoting Zhang, Wei Liu, Dimitris N. Metaxas
Abstract	In this paper, we propose a novel visual tracking framework that intelligently discovers reliable patterns from a wide range of video to resist drift error for long-term tracking tasks. First, we design a Discrete Fourier Transform (DFT) based tracker which is able to exploit a large number of tracked samples while still ensures real-time performance. Second, we propose a clustering method with temporal constraints to explore and memorize consistent patterns from previous frames, named as reliable memories. By virtue of this method, our tracker can utilize uncontaminated information to alleviate drifting issues. Experimental results show that our tracker performs favorably against other state of-the-art methods on benchmark datasets. Furthermore, it is significantly competent in handling drifts and able to robustly track challenging long videos over 4000 frames, while most of others lose track at early frames.
Tasks	Visual Tracking
Published	2016-02-04
URL	http://arxiv.org/abs/1602.01887v2
PDF	http://arxiv.org/pdf/1602.01887v2.pdf
PWC	https://paperswithcode.com/paper/visual-tracking-via-reliable-memories
Repo
Framework

Differences between Industrial Models of Autonomy and Systemic Models of Autonomy


Title	Differences between Industrial Models of Autonomy and Systemic Models of Autonomy
Authors	Aleksander Lodwich
Abstract	This paper discusses the idea of levels of autonomy of systems - be this technical or organic - and compares the insights with models employed by industries used to describe maturity and capability of their products.
Tasks
Published	2016-05-24
URL	http://arxiv.org/abs/1605.07335v3
PDF	http://arxiv.org/pdf/1605.07335v3.pdf
PWC	https://paperswithcode.com/paper/differences-between-industrial-models-of
Repo
Framework

PixelNet: Towards a General Pixel-level Architecture


Title	PixelNet: Towards a General Pixel-level Architecture
Authors	Aayush Bansal, Xinlei Chen, Bryan Russell, Abhinav Gupta, Deva Ramanan
Abstract	We explore architectures for general pixel-level prediction problems, from low-level edge detection to mid-level surface normal estimation to high-level semantic segmentation. Convolutional predictors, such as the fully-convolutional network (FCN), have achieved remarkable success by exploiting the spatial redundancy of neighboring pixels through convolutional processing. Though computationally efficient, we point out that such approaches are not statistically efficient during learning precisely because spatial redundancy limits the information learned from neighboring pixels. We demonstrate that (1) stratified sampling allows us to add diversity during batch updates and (2) sampled multi-scale features allow us to explore more nonlinear predictors (multiple fully-connected layers followed by ReLU) that improve overall accuracy. Finally, our objective is to show how a architecture can get performance better than (or comparable to) the architectures designed for a particular task. Interestingly, our single architecture produces state-of-the-art results for semantic segmentation on PASCAL-Context, surface normal estimation on NYUDv2 dataset, and edge detection on BSDS without contextual post-processing.
Tasks	Edge Detection, Semantic Segmentation
Published	2016-09-21
URL	http://arxiv.org/abs/1609.06694v1
PDF	http://arxiv.org/pdf/1609.06694v1.pdf
PWC	https://paperswithcode.com/paper/pixelnet-towards-a-general-pixel-level
Repo
Framework

Recursive Regression with Neural Networks: Approximating the HJI PDE Solution


Title	Recursive Regression with Neural Networks: Approximating the HJI PDE Solution
Authors	Vicenç Rubies Royo, Claire Tomlin
Abstract	The majority of methods used to compute approximations to the Hamilton-Jacobi-Isaacs partial differential equation (HJI PDE) rely on the discretization of the state space to perform dynamic programming updates. This type of approach is known to suffer from the curse of dimensionality due to the exponential growth in grid points with the state dimension. In this work we present an approximate dynamic programming algorithm that computes an approximation of the solution of the HJI PDE by alternating between solving a regression problem and solving a minimax problem using a feedforward neural network as the function approximator. We find that this method requires less memory to run and to store the approximation than traditional gridding methods, and we test it on a few systems of two, three and six dimensions.
Tasks
Published	2016-11-08
URL	http://arxiv.org/abs/1611.02739v4
PDF	http://arxiv.org/pdf/1611.02739v4.pdf
PWC	https://paperswithcode.com/paper/recursive-regression-with-neural-networks
Repo
Framework

OC16-CE80: A Chinese-English Mixlingual Database and A Speech Recognition Baseline


Title	OC16-CE80: A Chinese-English Mixlingual Database and A Speech Recognition Baseline
Authors	Dong Wang, Zhiyuan Tang, Difei Tang, Qing Chen
Abstract	We present the OC16-CE80 Chinese-English mixlingual speech database which was released as a main resource for training, development and test for the Chinese-English mixlingual speech recognition (MixASR-CHEN) challenge on O-COCOSDA 2016. This database consists of 80 hours of speech signals recorded from more than 1,400 speakers, where the utterances are in Chinese but each involves one or several English words. Based on the database and another two free data resources (THCHS30 and the CMU dictionary), a speech recognition (ASR) baseline was constructed with the deep neural network-hidden Markov model (DNN-HMM) hybrid system. We then report the baseline results following the MixASR-CHEN evaluation rules and demonstrate that OC16-CE80 is a reasonable data resource for mixlingual research.
Tasks	Speech Recognition
Published	2016-09-27
URL	http://arxiv.org/abs/1609.08412v1
PDF	http://arxiv.org/pdf/1609.08412v1.pdf
PWC	https://paperswithcode.com/paper/oc16-ce80-a-chinese-english-mixlingual
Repo
Framework

Navigational Rule Derivation: An algorithm to determine the effect of traffic signs on road networks


Title	Navigational Rule Derivation: An algorithm to determine the effect of traffic signs on road networks
Authors	Daniil Galaktionov, Miguel R. Luaces, Ángeles S. Places
Abstract	In this paper we present an algorithm to build a road network map enriched with traffic rules such as one-way streets and forbidden turns, based on the interpretation of already detected and classified traffic signs. Such algorithm helps to automatize the elaboration of maps for commercial navigation systems. Our solution is based on simulating navigation along the road network, determining at each point of interest the visibility of the signs and their effect on the roads. We test our approach in a small urban network and discuss various ways to generalize it to support more complex environments.
Tasks
Published	2016-11-17
URL	http://arxiv.org/abs/1611.06108v1
PDF	http://arxiv.org/pdf/1611.06108v1.pdf
PWC	https://paperswithcode.com/paper/navigational-rule-derivation-an-algorithm-to
Repo
Framework

A General Retraining Framework for Scalable Adversarial Classification


Title	A General Retraining Framework for Scalable Adversarial Classification
Authors	Bo Li, Yevgeniy Vorobeychik, Xinyun Chen
Abstract	Traditional classification algorithms assume that training and test data come from similar distributions. This assumption is violated in adversarial settings, where malicious actors modify instances to evade detection. A number of custom methods have been developed for both adversarial evasion attacks and robust learning. We propose the first systematic and general-purpose retraining framework which can: a) boost robustness of an \emph{arbitrary} learning algorithm, in the face of b) a broader class of adversarial models than any prior methods. We show that, under natural conditions, the retraining framework minimizes an upper bound on optimal adversarial risk, and show how to extend this result to account for approximations of evasion attacks. Extensive experimental evaluation demonstrates that our retraining methods are nearly indistinguishable from state-of-the-art algorithms for optimizing adversarial risk, but are more general and far more scalable. The experiments also confirm that without retraining, our adversarial framework dramatically reduces the effectiveness of learning. In contrast, retraining significantly boosts robustness to evasion attacks without significantly compromising overall accuracy.
Tasks
Published	2016-04-09
URL	http://arxiv.org/abs/1604.02606v2
PDF	http://arxiv.org/pdf/1604.02606v2.pdf
PWC	https://paperswithcode.com/paper/a-general-retraining-framework-for-scalable
Repo
Framework

Finite Sample Analysis of Approximate Message Passing Algorithms


Title	Finite Sample Analysis of Approximate Message Passing Algorithms
Authors	Cynthia Rush, Ramji Venkataramanan
Abstract	Approximate message passing (AMP) refers to a class of efficient algorithms for statistical estimation in high-dimensional problems such as compressed sensing and low-rank matrix estimation. This paper analyzes the performance of AMP in the regime where the problem dimension is large but finite. For concreteness, we consider the setting of high-dimensional regression, where the goal is to estimate a high-dimensional vector $\beta_0$ from a noisy measurement $y=A \beta_0 + w$. AMP is a low-complexity, scalable algorithm for this problem. Under suitable assumptions on the measurement matrix $A$, AMP has the attractive feature that its performance can be accurately characterized in the large system limit by a simple scalar iteration called state evolution. Previous proofs of the validity of state evolution have all been asymptotic convergence results. In this paper, we derive a concentration inequality for AMP with i.i.d. Gaussian measurement matrices with finite size $n \times N$. The result shows that the probability of deviation from the state evolution prediction falls exponentially in $n$. This provides theoretical support for empirical findings that have demonstrated excellent agreement of AMP performance with state evolution predictions for moderately large dimensions. The concentration inequality also indicates that the number of AMP iterations $t$ can grow no faster than order $\frac{\log n}{\log \log n}$ for the performance to be close to the state evolution predictions with high probability. The analysis can be extended to obtain similar non-asymptotic results for AMP in other settings such as low-rank matrix estimation.
Tasks
Published	2016-06-06
URL	http://arxiv.org/abs/1606.01800v4
PDF	http://arxiv.org/pdf/1606.01800v4.pdf
PWC	https://paperswithcode.com/paper/finite-sample-analysis-of-approximate-message
Repo
Framework