Paper Group ANR 1631
Chainer: A Deep Learning Framework for Accelerating the Research Cycle. Multi-sense Definition Modeling using Word Sense Decompositions. Matching Thermal to Visible Face Images Using a Semantic-Guided Generative Adversarial Network. Black-Box Decision based Adversarial Attack with Symmetric $α$-stable Distribution. Domain-Specific Priors and Meta L …
Chainer: A Deep Learning Framework for Accelerating the Research Cycle
| Title | Chainer: A Deep Learning Framework for Accelerating the Research Cycle |
| Authors | Seiya Tokui, Ryosuke Okuta, Takuya Akiba, Yusuke Niitani, Toru Ogawa, Shunta Saito, Shuji Suzuki, Kota Uenishi, Brian Vogel, Hiroyuki Yamazaki Vincent |
| Abstract | Software frameworks for neural networks play a key role in the development and application of deep learning methods. In this paper, we introduce the Chainer framework, which intends to provide a flexible, intuitive, and high performance means of implementing the full range of deep learning models needed by researchers and practitioners. Chainer provides acceleration using Graphics Processing Units with a familiar NumPy-like API through CuPy, supports general and dynamic models in Python through Define-by-Run, and also provides add-on packages for state-of-the-art computer vision models as well as distributed training. |
| Tasks | |
| Published | 2019-08-01 |
| URL | https://arxiv.org/abs/1908.00213v1 |
| https://arxiv.org/pdf/1908.00213v1.pdf | |
| PWC | https://paperswithcode.com/paper/chainer-a-deep-learning-framework-for |
| Repo | |
| Framework | |
Multi-sense Definition Modeling using Word Sense Decompositions
| Title | Multi-sense Definition Modeling using Word Sense Decompositions |
| Authors | Ruimin Zhu, Thanapon Noraset, Alisa Liu, Wenxin Jiang, Doug Downey |
| Abstract | Word embeddings capture syntactic and semantic information about words. Definition modeling aims to make the semantic content in each embedding explicit, by outputting a natural language definition based on the embedding. However, existing definition models are limited in their ability to generate accurate definitions for different senses of the same word. In this paper, we introduce a new method that enables definition modeling for multiple senses. We show how a Gumble-Softmax approach outperforms baselines at matching sense-specific embeddings to definitions during training. In experiments, our multi-sense definition model improves recall over a state-of-the-art single-sense definition model by a factor of three, without harming precision. |
| Tasks | Word Embeddings |
| Published | 2019-09-19 |
| URL | https://arxiv.org/abs/1909.09483v1 |
| https://arxiv.org/pdf/1909.09483v1.pdf | |
| PWC | https://paperswithcode.com/paper/multi-sense-definition-modeling-using-word |
| Repo | |
| Framework | |
Matching Thermal to Visible Face Images Using a Semantic-Guided Generative Adversarial Network
| Title | Matching Thermal to Visible Face Images Using a Semantic-Guided Generative Adversarial Network |
| Authors | Cunjian Chen, Arun Ross |
| Abstract | Designing face recognition systems that are capable of matching face images obtained in the thermal spectrum with those obtained in the visible spectrum is a challenging problem. In this work, we propose the use of semantic-guided generative adversarial network (SG-GAN) to automatically synthesize visible face images from their thermal counterparts. Specifically, semantic labels, extracted by a face parsing network, are used to compute a semantic loss function to regularize the adversarial network during training. These semantic cues denote high-level facial component information associated with each pixel. Further, an identity extraction network is leveraged to generate multi-scale features to compute an identity loss function. To achieve photo-realistic results, a perceptual loss function is introduced during network training to ensure that the synthesized visible face is perceptually similar to the target visible face image. We extensively evaluate the benefits of individual loss functions, and combine them effectively to learn the mapping from thermal to visible face images. Experiments involving two multispectral face datasets show that the proposed method achieves promising results in both face synthesis and cross-spectral face matching. |
| Tasks | Face Generation, Face Recognition |
| Published | 2019-03-03 |
| URL | http://arxiv.org/abs/1903.00963v1 |
| http://arxiv.org/pdf/1903.00963v1.pdf | |
| PWC | https://paperswithcode.com/paper/matching-thermal-to-visible-face-images-using |
| Repo | |
| Framework | |
Black-Box Decision based Adversarial Attack with Symmetric $α$-stable Distribution
| Title | Black-Box Decision based Adversarial Attack with Symmetric $α$-stable Distribution |
| Authors | Vignesh Srinivasan, Ercan E. Kuruoglu, Klaus-Robert Müller, Wojciech Samek, Shinichi Nakajima |
| Abstract | Developing techniques for adversarial attack and defense is an important research field for establishing reliable machine learning and its applications. Many existing methods employ Gaussian random variables for exploring the data space to find the most adversarial (for attacking) or least adversarial (for defense) point. However, the Gaussian distribution is not necessarily the optimal choice when the exploration is required to follow the complicated structure that most real-world data distributions exhibit. In this paper, we investigate how statistics of random variables affect such random walk exploration. Specifically, we generalize the Boundary Attack, a state-of-the-art black-box decision based attacking strategy, and propose the L'evy-Attack, where the random walk is driven by symmetric $\alpha$-stable random variables. Our experiments on MNIST and CIFAR10 datasets show that the L'evy-Attack explores the image data space more efficiently, and significantly improves the performance. Our results also give an insight into the recently found fact in the whitebox attacking scenario that the choice of the norm for measuring the amplitude of the adversarial patterns is essential. |
| Tasks | Adversarial Attack |
| Published | 2019-04-11 |
| URL | http://arxiv.org/abs/1904.05586v1 |
| http://arxiv.org/pdf/1904.05586v1.pdf | |
| PWC | https://paperswithcode.com/paper/black-box-decision-based-adversarial-attack |
| Repo | |
| Framework | |
Domain-Specific Priors and Meta Learning for Low-shot First-Person Action Recognition
| Title | Domain-Specific Priors and Meta Learning for Low-shot First-Person Action Recognition |
| Authors | Huseyin Coskun, Zeeshan Zia, Bugra Tekin, Federica Bogo, Nassir Navab, Federico Tombari, Harpreet Sawhney |
| Abstract | The lack of large-scale real datasets with annotationsmakes transfer learning a necessity for video activity under-standing. Within this scope, we aim at developing an effec-tive method for low-shot transfer learning for first-personaction classification. We leverage independently trained lo-cal visual cues to learn representations that can be trans-ferred from a source domain providing primitive action la-bels to a target domain with only a handful of examples.Such visual cues include object-object interactions, handgrasps and motion within regions that are a function of handlocations. We suggest a framework based on meta-learningto appropriately extract the distinctive and domain invari-ant components of the deployed visual cues, so to be able totransfer action classification models across public datasetscaptured with different scene configurations. We thoroughlyevaluate our methodology and report promising results overstate-of-the-art action classification approaches for bothinter-class and inter-dataset transfer. |
| Tasks | Action Classification, Meta-Learning, Transfer Learning |
| Published | 2019-07-22 |
| URL | https://arxiv.org/abs/1907.09382v1 |
| https://arxiv.org/pdf/1907.09382v1.pdf | |
| PWC | https://paperswithcode.com/paper/domain-specific-priors-and-meta-learning-for |
| Repo | |
| Framework | |
Selection-based Question Answering of an MOOC
| Title | Selection-based Question Answering of an MOOC |
| Authors | Atul Sahay, Smita Gholkar, Kavi Arya |
| Abstract | e-Yantra Robotics Competition (eYRC) is a unique Robotics Competition hosted by IIT Bombay that is actually an Embedded Systems and Robotics MOOC. Registrations have been growing exponentially in each year from 4500 in 2012 to over 34000 in 2019. In this 5-month long competition students learn complex skills under severe time pressure and have access to a discussion forum to post doubts about the learning material. Responding to questions in real-time is a challenge for project staff. Here, we illustrate the advantage of Deep Learning for real-time question answering in the eYRC discussion forum. We illustrate the advantage of Transformer based contextual embedding mechanisms such as Bidirectional Encoder Representation From Transformer (BERT) over word embedding mechanisms such as Word2Vec. We propose a weighted similarity metric as a measure of matching and find it more reliable than Content-Content or Title-Title similarities alone. The automation of replying to questions has brought the turn around response time(TART) down from a minimum of 21 mins to a minimum of 0.3 secs. |
| Tasks | Question Answering |
| Published | 2019-11-15 |
| URL | https://arxiv.org/abs/1911.07629v1 |
| https://arxiv.org/pdf/1911.07629v1.pdf | |
| PWC | https://paperswithcode.com/paper/selection-based-question-answering-of-an-mooc |
| Repo | |
| Framework | |
Augmented Memory Networks for Streaming-Based Active One-Shot Learning
| Title | Augmented Memory Networks for Streaming-Based Active One-Shot Learning |
| Authors | Andreas Kvistad, Massimiliano Ruocco, Eliezer de Souza da Silva, Erlend Aune |
| Abstract | One of the major challenges in training deep architectures for predictive tasks is the scarcity and cost of labeled training data. Active Learning (AL) is one way of addressing this challenge. In stream-based AL, observations are continuously made available to the learner that have to decide whether to request a label or to make a prediction. The goal is to reduce the request rate while at the same time maximize prediction performance. In previous research, reinforcement learning has been used for learning the AL request/prediction strategy. In our work, we propose to equip a reinforcement learning process with memory augmented neural networks, to enhance the one-shot capabilities. Moreover, we introduce Class Margin Sampling (CMS) as an extension of the standard margin sampling to the reinforcement learning setting. This strategy aims to reduce training time and improve sample efficiency in the training process. We evaluate the proposed method on a classification task using empirical accuracy of label predictions and percentage of label requests. The results indicates that the proposed method, by making use of the memory augmented networks and CMS in the training process, outperforms existing baselines. |
| Tasks | Active Learning, One-Shot Learning |
| Published | 2019-09-04 |
| URL | https://arxiv.org/abs/1909.01757v1 |
| https://arxiv.org/pdf/1909.01757v1.pdf | |
| PWC | https://paperswithcode.com/paper/augmented-memory-networks-for-streaming-based |
| Repo | |
| Framework | |
General Fragment Model for Information Artifacts
| Title | General Fragment Model for Information Artifacts |
| Authors | Sandro Rama Fiorini, Wallas Sousa dos Santos, Rodrigo Costa Mesquita, Guilherme Ferreira Lima, Marcio F. Moreno |
| Abstract | The use of semantic descriptions in data intensive domains require a systematic model for linking semantic descriptions with their manifestations in fragments of heterogeneous information and data objects. Such information heterogeneity requires a fragment model that is general enough to support the specification of anchors from conceptual models to multiple types of information artifacts. While diverse proposals of anchoring models exist in the literature, they are usually focused in audiovisual information. We propose a generalized fragment model that can be instantiated to different kinds of information artifacts. Our objective is to systematize the way in which fragments and anchors can be described in conceptual models, without committing to a specific vocabulary. |
| Tasks | |
| Published | 2019-09-09 |
| URL | https://arxiv.org/abs/1909.04117v1 |
| https://arxiv.org/pdf/1909.04117v1.pdf | |
| PWC | https://paperswithcode.com/paper/general-fragment-model-for-information |
| Repo | |
| Framework | |
SPM-Tracker: Series-Parallel Matching for Real-Time Visual Object Tracking
| Title | SPM-Tracker: Series-Parallel Matching for Real-Time Visual Object Tracking |
| Authors | Guangting Wang, Chong Luo, Zhiwei Xiong, Wenjun Zeng |
| Abstract | The greatest challenge facing visual object tracking is the simultaneous requirements on robustness and discrimination power. In this paper, we propose a SiamFC-based tracker, named SPM-Tracker, to tackle this challenge. The basic idea is to address the two requirements in two separate matching stages. Robustness is strengthened in the coarse matching (CM) stage through generalized training while discrimination power is enhanced in the fine matching (FM) stage through a distance learning network. The two stages are connected in series as the input proposals of the FM stage are generated by the CM stage. They are also connected in parallel as the matching scores and box location refinements are fused to generate the final results. This innovative series-parallel structure takes advantage of both stages and results in superior performance. The proposed SPM-Tracker, running at 120fps on GPU, achieves an AUC of 0.687 on OTB-100 and an EAO of 0.434 on VOT-16, exceeding other real-time trackers by a notable margin. |
| Tasks | Object Tracking, Visual Object Tracking |
| Published | 2019-04-09 |
| URL | http://arxiv.org/abs/1904.04452v1 |
| http://arxiv.org/pdf/1904.04452v1.pdf | |
| PWC | https://paperswithcode.com/paper/spm-tracker-series-parallel-matching-for-real |
| Repo | |
| Framework | |
Unsupervised Ensemble Classification with Dependent Data
| Title | Unsupervised Ensemble Classification with Dependent Data |
| Authors | Panagiotis A. Traganitis, Georgios B. Giannakis |
| Abstract | Ensemble learning, the machine learning paradigm where multiple algorithms are combined, has exhibited promising perfomance in a variety of tasks. The present work focuses on unsupervised ensemble classification. The term unsupervised refers to the ensemble combiner who has no knowledge of the ground-truth labels that each classifier has been trained on. While most prior works on unsupervised ensemble classification are designed for independent and identically distributed (i.i.d.) data, the present work introduces an unsupervised scheme for learning from ensembles of classifiers in the presence of data dependencies. Two types of data dependencies are considered: sequential data and networked data whose dependencies are captured by a graph. Moment matching and Expectation Maximization algorithms are developed for the aforementioned cases, and their performance is evaluated on synthetic and real datasets. |
| Tasks | |
| Published | 2019-06-22 |
| URL | https://arxiv.org/abs/1906.09356v1 |
| https://arxiv.org/pdf/1906.09356v1.pdf | |
| PWC | https://paperswithcode.com/paper/unsupervised-ensemble-classification-with |
| Repo | |
| Framework | |
Margin-Based Generalization Lower Bounds for Boosted Classifiers
| Title | Margin-Based Generalization Lower Bounds for Boosted Classifiers |
| Authors | Allan Grønlund, Lior Kamma, Kasper Green Larsen, Alexander Mathiasen, Jelani Nelson |
| Abstract | Boosting is one of the most successful ideas in machine learning. The most well-accepted explanations for the low generalization error of boosting algorithms such as AdaBoost stem from margin theory. The study of margins in the context of boosting algorithms was initiated by Schapire, Freund, Bartlett and Lee (1998) and has inspired numerous boosting algorithms and generalization bounds. To date, the strongest known generalization (upper bound) is the $k$th margin bound of Gao and Zhou (2013). Despite the numerous generalization upper bounds that have been proved over the last two decades, nothing is known about the tightness of these bounds. In this paper, we give the first margin-based lower bounds on the generalization error of boosted classifiers. Our lower bounds nearly match the $k$th margin bound and thus almost settle the generalization performance of boosted classifiers in terms of margins. |
| Tasks | |
| Published | 2019-09-27 |
| URL | https://arxiv.org/abs/1909.12518v2 |
| https://arxiv.org/pdf/1909.12518v2.pdf | |
| PWC | https://paperswithcode.com/paper/margin-based-generalization-lower-bounds-for |
| Repo | |
| Framework | |
Detection of the Group of Traffic Signs with Central Slice Theorem
| Title | Detection of the Group of Traffic Signs with Central Slice Theorem |
| Authors | Koba Natroshvili |
| Abstract | Our sensor system consists of a combination of Photonic Mixer Device - PMD and Mono optical cameras. Some traffic signs have stripes at 45{deg}. These traffic signs cancel different restrictions on the road. We detect this class of signs with Radon transformation. Here the Radon transformation is calculated using Central Slice Theorem. We approximate the slice of spectrum by the Discrete Cosine Transformation (DCT). |
| Tasks | |
| Published | 2019-08-04 |
| URL | https://arxiv.org/abs/1908.04386v1 |
| https://arxiv.org/pdf/1908.04386v1.pdf | |
| PWC | https://paperswithcode.com/paper/detection-of-the-group-of-traffic-signs-with |
| Repo | |
| Framework | |
On the Robustness of Data-Driven Controllers for Linear Systems
| Title | On the Robustness of Data-Driven Controllers for Linear Systems |
| Authors | Rajasekhar Anguluri, Abed AlRahman Al Makdah, Vaibhav Katewa, Fabio Pasqualetti |
| Abstract | This paper proposes a new framework and several results to quantify the performance of data-driven state-feedback controllers for linear systems against targeted perturbations of the training data. We focus on the case where subsets of the training data are randomly corrupted by an adversary, and derive lower and upper bounds for the stability of the closed-loop system with compromised controller as a function of the perturbation statistics, size of the training data, sensitivity of the data-driven algorithm to perturbation of the training data, and properties of the nominal closed-loop system. Our stability and convergence bounds are probabilistic in nature, and rely on a first-order approximation of the data-driven procedure that designs the state-feedback controller, which can be computed directly using the training data. We illustrate our findings via multiple numerical studies. |
| Tasks | |
| Published | 2019-12-21 |
| URL | https://arxiv.org/abs/1912.10231v1 |
| https://arxiv.org/pdf/1912.10231v1.pdf | |
| PWC | https://paperswithcode.com/paper/on-the-robustness-of-data-driven-controllers |
| Repo | |
| Framework | |
A Concept-Value Network as a Brain Model
| Title | A Concept-Value Network as a Brain Model |
| Authors | Kieran Greer |
| Abstract | This paper suggests a statistical framework for describing the relations between the physical and conceptual entities of a brain-like model. In particular, features and concept instances are put into context. This may help with understanding or implementing a similar model. The paper suggests that features are in fact the wiring. With this idea, the actual length of the connection is important, because it is related to firing rates and neuron synchronization. The paper then suggests that concepts are neuron groups that link features and concept instances are the signals from those groups. Therefore, features become the static framework of the interconnected neural system and concepts are combinations of these, as determined by the external stimulus and the neural synaptic strengths. Along with this statistical model, it is possible to propose a simplified design for a neuron, based on an action potential and variable output signal. A strong comparison with Hebbian theory is then proposed, with some test results to support the theory. |
| Tasks | |
| Published | 2019-04-09 |
| URL | https://arxiv.org/abs/1904.04579v2 |
| https://arxiv.org/pdf/1904.04579v2.pdf | |
| PWC | https://paperswithcode.com/paper/a-feature-value-network-as-a-brain-model |
| Repo | |
| Framework | |
Quant GANs: Deep Generation of Financial Time Series
| Title | Quant GANs: Deep Generation of Financial Time Series |
| Authors | Magnus Wiese, Robert Knobloch, Ralf Korn, Peter Kretschmer |
| Abstract | Modeling financial time series by stochastic processes is a challenging task and a central area of research in financial mathematics. As an alternative, we introduce Quant GANs, a data-driven model which is inspired by the recent success of generative adversarial networks (GANs). Quant GANs consist of a generator and discriminator function, which utilize temporal convolutional networks (TCNs) and thereby achieve to capture long-range dependencies such as the presence of volatility clusters. The generator function is explicitly constructed such that the induced stochastic process allows a transition to its risk-neutral distribution. Our numerical results highlight that distributional properties for small and large lags are in an excellent agreement and dependence properties such as volatility clusters, leverage effects, and serial autocorrelations can be generated by the generator function of Quant GANs, demonstrably in high fidelity. |
| Tasks | Time Series |
| Published | 2019-07-15 |
| URL | https://arxiv.org/abs/1907.06673v2 |
| https://arxiv.org/pdf/1907.06673v2.pdf | |
| PWC | https://paperswithcode.com/paper/quant-gans-deep-generation-of-financial-time |
| Repo | |
| Framework | |
