Paper Group AWR 11
Shallow and Deep Convolutional Networks for Saliency Prediction. Temporal Action Localization in Untrimmed Videos via Multi-stage CNNs. An Efficient Training Algorithm for Kernel Survival Support Vector Machines. Double Thompson Sampling for Dueling Bandits. Temporal Activity Detection in Untrimmed Videos with Recurrent Neural Networks. Causal Lear …
Shallow and Deep Convolutional Networks for Saliency Prediction
| Title | Shallow and Deep Convolutional Networks for Saliency Prediction |
| Authors | Junting Pan, Kevin McGuinness, Elisa Sayrol, Noel O’Connor, Xavier Giro-i-Nieto |
| Abstract | The prediction of salient areas in images has been traditionally addressed with hand-crafted features based on neuroscience principles. This paper, however, addresses the problem with a completely data-driven approach by training a convolutional neural network (convnet). The learning process is formulated as a minimization of a loss function that measures the Euclidean distance of the predicted saliency map with the provided ground truth. The recent publication of large datasets of saliency prediction has provided enough data to train end-to-end architectures that are both fast and accurate. Two designs are proposed: a shallow convnet trained from scratch, and a another deeper solution whose first three layers are adapted from another network trained for classification. To the authors knowledge, these are the first end-to-end CNNs trained and tested for the purpose of saliency prediction. |
| Tasks | Saliency Prediction |
| Published | 2016-03-02 |
| URL | http://arxiv.org/abs/1603.00845v1 |
| http://arxiv.org/pdf/1603.00845v1.pdf | |
| PWC | https://paperswithcode.com/paper/shallow-and-deep-convolutional-networks-for |
| Repo | https://github.com/imatge-upc/saliency-2016-cvpr |
| Framework | none |
Temporal Action Localization in Untrimmed Videos via Multi-stage CNNs
| Title | Temporal Action Localization in Untrimmed Videos via Multi-stage CNNs |
| Authors | Zheng Shou, Dongang Wang, Shih-Fu Chang |
| Abstract | We address temporal action localization in untrimmed long videos. This is important because videos in real applications are usually unconstrained and contain multiple action instances plus video content of background scenes or other activities. To address this challenging issue, we exploit the effectiveness of deep networks in temporal action localization via three segment-based 3D ConvNets: (1) a proposal network identifies candidate segments in a long video that may contain actions; (2) a classification network learns one-vs-all action classification model to serve as initialization for the localization network; and (3) a localization network fine-tunes on the learned classification network to localize each action instance. We propose a novel loss function for the localization network to explicitly consider temporal overlap and therefore achieve high temporal localization accuracy. Only the proposal network and the localization network are used during prediction. On two large-scale benchmarks, our approach achieves significantly superior performances compared with other state-of-the-art systems: mAP increases from 1.7% to 7.4% on MEXaction2 and increases from 15.0% to 19.0% on THUMOS 2014, when the overlap threshold for evaluation is set to 0.5. |
| Tasks | Action Classification, Action Localization, Temporal Action Localization, Temporal Localization |
| Published | 2016-01-09 |
| URL | http://arxiv.org/abs/1601.02129v2 |
| http://arxiv.org/pdf/1601.02129v2.pdf | |
| PWC | https://paperswithcode.com/paper/temporal-action-localization-in-untrimmed |
| Repo | https://github.com/zhengshou/scnn |
| Framework | none |
An Efficient Training Algorithm for Kernel Survival Support Vector Machines
| Title | An Efficient Training Algorithm for Kernel Survival Support Vector Machines |
| Authors | Sebastian Pölsterl, Nassir Navab, Amin Katouzian |
| Abstract | Survival analysis is a fundamental tool in medical research to identify predictors of adverse events and develop systems for clinical decision support. In order to leverage large amounts of patient data, efficient optimisation routines are paramount. We propose an efficient training algorithm for the kernel survival support vector machine (SSVM). We directly optimise the primal objective function and employ truncated Newton optimisation and order statistic trees to significantly lower computational costs compared to previous training algorithms, which require $O(n^4)$ space and $O(p n^6)$ time for datasets with $n$ samples and $p$ features. Our results demonstrate that our proposed optimisation scheme allows analysing data of a much larger scale with no loss in prediction performance. Experiments on synthetic and 5 real-world datasets show that our technique outperforms existing kernel SSVM formulations if the amount of right censoring is high ($\geq85%$), and performs comparably otherwise. |
| Tasks | Survival Analysis |
| Published | 2016-11-21 |
| URL | http://arxiv.org/abs/1611.07054v1 |
| http://arxiv.org/pdf/1611.07054v1.pdf | |
| PWC | https://paperswithcode.com/paper/an-efficient-training-algorithm-for-kernel |
| Repo | https://github.com/sebp/scikit-survival |
| Framework | none |
Double Thompson Sampling for Dueling Bandits
| Title | Double Thompson Sampling for Dueling Bandits |
| Authors | Huasen Wu, Xin Liu |
| Abstract | In this paper, we propose a Double Thompson Sampling (D-TS) algorithm for dueling bandit problems. As indicated by its name, D-TS selects both the first and the second candidates according to Thompson Sampling. Specifically, D-TS maintains a posterior distribution for the preference matrix, and chooses the pair of arms for comparison by sampling twice from the posterior distribution. This simple algorithm applies to general Copeland dueling bandits, including Condorcet dueling bandits as its special case. For general Copeland dueling bandits, we show that D-TS achieves $O(K^2 \log T)$ regret. For Condorcet dueling bandits, we further simplify the D-TS algorithm and show that the simplified D-TS algorithm achieves $O(K \log T + K^2 \log \log T)$ regret. Simulation results based on both synthetic and real-world data demonstrate the efficiency of the proposed D-TS algorithm. |
| Tasks | |
| Published | 2016-04-25 |
| URL | http://arxiv.org/abs/1604.07101v2 |
| http://arxiv.org/pdf/1604.07101v2.pdf | |
| PWC | https://paperswithcode.com/paper/double-thompson-sampling-for-dueling-bandits |
| Repo | https://github.com/HuasenWu/DuelingBandits |
| Framework | none |
Temporal Activity Detection in Untrimmed Videos with Recurrent Neural Networks
| Title | Temporal Activity Detection in Untrimmed Videos with Recurrent Neural Networks |
| Authors | Alberto Montes, Amaia Salvador, Santiago Pascual, Xavier Giro-i-Nieto |
| Abstract | This thesis explore different approaches using Convolutional and Recurrent Neural Networks to classify and temporally localize activities on videos, furthermore an implementation to achieve it has been proposed. As the first step, features have been extracted from video frames using an state of the art 3D Convolutional Neural Network. This features are fed in a recurrent neural network that solves the activity classification and temporally location tasks in a simple and flexible way. Different architectures and configurations have been tested in order to achieve the best performance and learning of the video dataset provided. In addition it has been studied different kind of post processing over the trained network’s output to achieve a better results on the temporally localization of activities on the videos. The results provided by the neural network developed in this thesis have been submitted to the ActivityNet Challenge 2016 of the CVPR, achieving competitive results using a simple and flexible architecture. |
| Tasks | Action Detection, Activity Detection |
| Published | 2016-08-29 |
| URL | http://arxiv.org/abs/1608.08128v3 |
| http://arxiv.org/pdf/1608.08128v3.pdf | |
| PWC | https://paperswithcode.com/paper/temporal-activity-detection-in-untrimmed |
| Repo | https://github.com/vohoaiviet/activitynet-2016-cvprw |
| Framework | tf |
Causal Learning via Manifold Regularization
| Title | Causal Learning via Manifold Regularization |
| Authors | Steven M. Hill, Chris. J. Oates, Duncan A. Blythe, Sach Mukherjee |
| Abstract | This paper frames causal structure estimation as a machine learning task. The idea is to treat indicators of causal relationships between variables as `labels’ and to exploit available data on the variables of interest to provide features for the labelling task. Background scientific knowledge or any available interventional data provide labels on some causal relationships and the remainder are treated as unlabelled. To illustrate the key ideas, we develop a distance-based approach (based on bivariate histograms) within a manifold regularization framework. We present empirical results on three different biological data sets (including examples where causal effects can be verified by experimental intervention), that together demonstrate the efficacy and general nature of the approach as well as its simplicity from a user’s point of view. | |
| Tasks | Causal Discovery |
| Published | 2016-12-16 |
| URL | https://arxiv.org/abs/1612.05678v4 |
| https://arxiv.org/pdf/1612.05678v4.pdf | |
| PWC | https://paperswithcode.com/paper/causal-discovery-as-semi-supervised-learning |
| Repo | https://github.com/Steven-M-Hill/MRCL |
| Framework | none |
Equilibrium Propagation: Bridging the Gap Between Energy-Based Models and Backpropagation
| Title | Equilibrium Propagation: Bridging the Gap Between Energy-Based Models and Backpropagation |
| Authors | Benjamin Scellier, Yoshua Bengio |
| Abstract | We introduce Equilibrium Propagation, a learning framework for energy-based models. It involves only one kind of neural computation, performed in both the first phase (when the prediction is made) and the second phase of training (after the target or prediction error is revealed). Although this algorithm computes the gradient of an objective function just like Backpropagation, it does not need a special computation or circuit for the second phase, where errors are implicitly propagated. Equilibrium Propagation shares similarities with Contrastive Hebbian Learning and Contrastive Divergence while solving the theoretical issues of both algorithms: our algorithm computes the gradient of a well defined objective function. Because the objective function is defined in terms of local perturbations, the second phase of Equilibrium Propagation corresponds to only nudging the prediction (fixed point, or stationary distribution) towards a configuration that reduces prediction error. In the case of a recurrent multi-layer supervised network, the output units are slightly nudged towards their target in the second phase, and the perturbation introduced at the output layer propagates backward in the hidden layers. We show that the signal ‘back-propagated’ during this second phase corresponds to the propagation of error derivatives and encodes the gradient of the objective function, when the synaptic update corresponds to a standard form of spike-timing dependent plasticity. This work makes it more plausible that a mechanism similar to Backpropagation could be implemented by brains, since leaky integrator neural computation performs both inference and error back-propagation in our model. The only local difference between the two phases is whether synaptic changes are allowed or not. |
| Tasks | |
| Published | 2016-02-16 |
| URL | http://arxiv.org/abs/1602.05179v5 |
| http://arxiv.org/pdf/1602.05179v5.pdf | |
| PWC | https://paperswithcode.com/paper/equilibrium-propagation-bridging-the-gap |
| Repo | https://github.com/bscellier/Towards-a-Biologically-Plausible-Backprop |
| Framework | pytorch |
Ternary Weight Networks
| Title | Ternary Weight Networks |
| Authors | Fengfu Li, Bo Zhang, Bin Liu |
| Abstract | We introduce ternary weight networks (TWNs) - neural networks with weights constrained to +1, 0 and -1. The Euclidian distance between full (float or double) precision weights and the ternary weights along with a scaling factor is minimized. Besides, a threshold-based ternary function is optimized to get an approximated solution which can be fast and easily computed. TWNs have stronger expressive abilities than the recently proposed binary precision counterparts and are thus more effective than the latter. Meanwhile, TWNs achieve up to 16$\times$ or 32$\times$ model compression rate and need fewer multiplications compared with the full precision counterparts. Benchmarks on MNIST, CIFAR-10, and large scale ImageNet datasets show that the performance of TWNs is only slightly worse than the full precision counterparts but outperforms the analogous binary precision counterparts a lot. |
| Tasks | Model Compression |
| Published | 2016-05-16 |
| URL | http://arxiv.org/abs/1605.04711v2 |
| http://arxiv.org/pdf/1605.04711v2.pdf | |
| PWC | https://paperswithcode.com/paper/ternary-weight-networks |
| Repo | https://github.com/da-steve101/binary_connect_cifar |
| Framework | none |
Minimizing Quadratic Functions in Constant Time
| Title | Minimizing Quadratic Functions in Constant Time |
| Authors | Kohei Hayashi, Yuichi Yoshida |
| Abstract | A sampling-based optimization method for quadratic functions is proposed. Our method approximately solves the following $n$-dimensional quadratic minimization problem in constant time, which is independent of $n$: $z^*=\min_{\mathbf{v} \in \mathbb{R}^n}\langle\mathbf{v}, A \mathbf{v}\rangle + n\langle\mathbf{v}, \mathrm{diag}(\mathbf{d})\mathbf{v}\rangle + n\langle\mathbf{b}, \mathbf{v}\rangle$, where $A \in \mathbb{R}^{n \times n}$ is a matrix and $\mathbf{d},\mathbf{b} \in \mathbb{R}^n$ are vectors. Our theoretical analysis specifies the number of samples $k(\delta, \epsilon)$ such that the approximated solution $z$ satisfies $z - z^* = O(\epsilon n^2)$ with probability $1-\delta$. The empirical performance (accuracy and runtime) is positively confirmed by numerical experiments. |
| Tasks | |
| Published | 2016-08-25 |
| URL | http://arxiv.org/abs/1608.07179v1 |
| http://arxiv.org/pdf/1608.07179v1.pdf | |
| PWC | https://paperswithcode.com/paper/minimizing-quadratic-functions-in-constant |
| Repo | https://github.com/hayasick/CTOQ |
| Framework | none |
Adversarial Training Methods for Semi-Supervised Text Classification
| Title | Adversarial Training Methods for Semi-Supervised Text Classification |
| Authors | Takeru Miyato, Andrew M. Dai, Ian Goodfellow |
| Abstract | Adversarial training provides a means of regularizing supervised learning algorithms while virtual adversarial training is able to extend supervised learning algorithms to the semi-supervised setting. However, both methods require making small perturbations to numerous entries of the input vector, which is inappropriate for sparse high-dimensional inputs such as one-hot word representations. We extend adversarial and virtual adversarial training to the text domain by applying perturbations to the word embeddings in a recurrent neural network rather than to the original input itself. The proposed method achieves state of the art results on multiple benchmark semi-supervised and purely supervised tasks. We provide visualizations and analysis showing that the learned word embeddings have improved in quality and that while training, the model is less prone to overfitting. |
| Tasks | Sentiment Analysis, Text Classification, Word Embeddings |
| Published | 2016-05-25 |
| URL | http://arxiv.org/abs/1605.07725v3 |
| http://arxiv.org/pdf/1605.07725v3.pdf | |
| PWC | https://paperswithcode.com/paper/adversarial-training-methods-for-semi |
| Repo | https://github.com/TobiasLee/Text-Classification |
| Framework | tf |
Relation Schema Induction using Tensor Factorization with Side Information
| Title | Relation Schema Induction using Tensor Factorization with Side Information |
| Authors | Madhav Nimishakavi, Uday Singh Saini, Partha Talukdar |
| Abstract | Given a set of documents from a specific domain (e.g., medical research journals), how do we automatically build a Knowledge Graph (KG) for that domain? Automatic identification of relations and their schemas, i.e., type signature of arguments of relations (e.g., undergo(Patient, Surgery)), is an important first step towards this goal. We refer to this problem as Relation Schema Induction (RSI). In this paper, we propose Schema Induction using Coupled Tensor Factorization (SICTF), a novel tensor factorization method for relation schema induction. SICTF factorizes Open Information Extraction (OpenIE) triples extracted from a domain corpus along with additional side information in a principled way to induce relation schemas. To the best of our knowledge, this is the first application of tensor factorization for the RSI problem. Through extensive experiments on multiple real-world datasets, we find that SICTF is not only more accurate than state-of-the-art baselines, but also significantly faster (about 14x faster). |
| Tasks | Open Information Extraction |
| Published | 2016-05-12 |
| URL | http://arxiv.org/abs/1605.04227v3 |
| http://arxiv.org/pdf/1605.04227v3.pdf | |
| PWC | https://paperswithcode.com/paper/relation-schema-induction-using-tensor |
| Repo | https://github.com/malllabiisc/sictf |
| Framework | none |
Quantum Algorithms for Compositional Natural Language Processing
| Title | Quantum Algorithms for Compositional Natural Language Processing |
| Authors | William Zeng, Bob Coecke |
| Abstract | We propose a new application of quantum computing to the field of natural language processing. Ongoing work in this field attempts to incorporate grammatical structure into algorithms that compute meaning. In (Coecke, Sadrzadeh and Clark, 2010), the authors introduce such a model (the CSC model) based on tensor product composition. While this algorithm has many advantages, its implementation is hampered by the large classical computational resources that it requires. In this work we show how computational shortcomings of the CSC approach could be resolved using quantum computation (possibly in addition to existing techniques for dimension reduction). We address the value of quantum RAM (Giovannetti,2008) for this model and extend an algorithm from Wiebe, Braun and Lloyd (2012) into a quantum algorithm to categorize sentences in CSC. Our new algorithm demonstrates a quadratic speedup over classical methods under certain conditions. |
| Tasks | Dimensionality Reduction |
| Published | 2016-08-04 |
| URL | http://arxiv.org/abs/1608.01406v1 |
| http://arxiv.org/pdf/1608.01406v1.pdf | |
| PWC | https://paperswithcode.com/paper/quantum-algorithms-for-compositional-natural |
| Repo | https://github.com/ICHEC/QNLP |
| Framework | none |
Multiple Object Tracking with Kernelized Correlation Filters in Urban Mixed Traffic
| Title | Multiple Object Tracking with Kernelized Correlation Filters in Urban Mixed Traffic |
| Authors | Yuebin Yang, Guillaume-Alexandre Bilodeau |
| Abstract | Recently, the Kernelized Correlation Filters tracker (KCF) achieved competitive performance and robustness in visual object tracking. On the other hand, visual trackers are not typically used in multiple object tracking. In this paper, we investigate how a robust visual tracker like KCF can improve multiple object tracking. Since KCF is a fast tracker, many can be used in parallel and still result in fast tracking. We build a multiple object tracking system based on KCF and background subtraction. Background subtraction is applied to extract moving objects and get their scale and size in combination with KCF outputs, while KCF is used for data association and to handle fragmentation and occlusion problems. As a result, KCF and background subtraction help each other to take tracking decision at every frame. Sometimes KCF outputs are the most trustworthy (e.g. during occlusion), while in some other case, it is the background subtraction outputs. To validate the effectiveness of our system, the algorithm is demonstrated on four urban video recordings from a standard dataset. Results show that our method is competitive with state-of-the-art trackers even if we use a much simpler data association step. |
| Tasks | Multiple Object Tracking, Object Tracking, Visual Object Tracking |
| Published | 2016-11-08 |
| URL | http://arxiv.org/abs/1611.02364v2 |
| http://arxiv.org/pdf/1611.02364v2.pdf | |
| PWC | https://paperswithcode.com/paper/multiple-object-tracking-with-kernelized |
| Repo | https://github.com/iyybpatrick/MKCF |
| Framework | none |
Fitting a Simplicial Complex using a Variation of k-means
| Title | Fitting a Simplicial Complex using a Variation of k-means |
| Authors | Piotr Beben |
| Abstract | We give a simple and effective two stage algorithm for approximating a point cloud $\mathcal{S}\subset\mathbb{R}^m$ by a simplicial complex $K$. The first stage is an iterative fitting procedure that generalizes k-means clustering, while the second stage involves deleting redundant simplices. A form of dimension reduction of $\mathcal{S}$ is obtained as a consequence. |
| Tasks | Dimensionality Reduction |
| Published | 2016-07-13 |
| URL | http://arxiv.org/abs/1607.03849v2 |
| http://arxiv.org/pdf/1607.03849v2.pdf | |
| PWC | https://paperswithcode.com/paper/fitting-a-simplicial-complex-using-a |
| Repo | https://github.com/pbebenSoton/smeans |
| Framework | none |
Character-Level Neural Translation for Multilingual Media Monitoring in the SUMMA Project
| Title | Character-Level Neural Translation for Multilingual Media Monitoring in the SUMMA Project |
| Authors | Guntis Barzdins, Steve Renals, Didzis Gosko |
| Abstract | The paper steps outside the comfort-zone of the traditional NLP tasks like automatic speech recognition (ASR) and machine translation (MT) to addresses two novel problems arising in the automated multilingual news monitoring: segmentation of the TV and radio program ASR transcripts into individual stories, and clustering of the individual stories coming from various sources and languages into storylines. Storyline clustering of stories covering the same events is an essential task for inquisitorial media monitoring. We address these two problems jointly by engaging the low-dimensional semantic representation capabilities of the sequence to sequence neural translation models. To enable joint multi-task learning for multilingual neural translation of morphologically rich languages we replace the attention mechanism with the sliding-window mechanism and operate the sequence to sequence neural translation model on the character-level rather than on the word-level. The story segmentation and storyline clustering problem is tackled by examining the low-dimensional vectors produced as a side-product of the neural translation process. The results of this paper describe a novel approach to the automatic story segmentation and storyline clustering problem. |
| Tasks | Machine Translation, Multi-Task Learning, Speech Recognition |
| Published | 2016-04-05 |
| URL | http://arxiv.org/abs/1604.01221v1 |
| http://arxiv.org/pdf/1604.01221v1.pdf | |
| PWC | https://paperswithcode.com/paper/character-level-neural-translation-for |
| Repo | https://github.com/didzis/tensorflowAMR |
| Framework | tf |
