Paper Group ANR 58
Inverse Visual Question Answering with Multi-Level Attentions. Matrix Cofactorization for Joint Representation Learning and Supervised Classification – Application to Hyperspectral Image Analysis. Differentiable probabilistic models of scientific imaging with the Fourier slice theorem. Federated Learning Of Out-Of-Vocabulary Words. A probabilistic …
Inverse Visual Question Answering with Multi-Level Attentions
| Title | Inverse Visual Question Answering with Multi-Level Attentions |
| Authors | Yaser Alwatter, Yuhong Guo |
| Abstract | In this paper, we propose a novel deep multi-level attention model to address inverse visual question answering. The proposed model generates regional visual and semantic features at the object level and then enhances them with the answer cue by using attention mechanisms. Two levels of multiple attentions are employed in the model, including the dual attention at the partial question encoding step and the dynamic attention at the next question word generation step. We evaluate the proposed model on the VQA V1 dataset. It demonstrates state-of-the-art performance in terms of multiple commonly used metrics. |
| Tasks | Question Answering, Visual Question Answering |
| Published | 2019-09-17 |
| URL | https://arxiv.org/abs/1909.07583v1 |
| https://arxiv.org/pdf/1909.07583v1.pdf | |
| PWC | https://paperswithcode.com/paper/inverse-visual-question-answering-with-multi |
| Repo | |
| Framework | |
Matrix Cofactorization for Joint Representation Learning and Supervised Classification – Application to Hyperspectral Image Analysis
| Title | Matrix Cofactorization for Joint Representation Learning and Supervised Classification – Application to Hyperspectral Image Analysis |
| Authors | Adrien Lagrange, Mathieu Fauvel, Stéphane May, José Bioucas-Dias, Nicolas Dobigeon |
| Abstract | Supervised classification and representation learning are two widely used classes of methods to analyze multivariate images. Although complementary, these methods have been scarcely considered jointly in a hierarchical modeling. In this paper, a method coupling these two approaches is designed using a matrix cofactorization formulation. Each task is modeled as a factorization matrix problem and a term relating both coding matrices is then introduced to drive an appropriate coupling. The link can be interpreted as a clustering operation over a low-dimensional representation vectors. The attribution vectors of the clustering are then used as features vectors for the classification task, i.e., the coding vectors of the corresponding factorization problem. A proximal gradient descent algorithm, ensuring convergence to a critical point of the objective function, is then derived to solve the resulting non-convex non-smooth optimization problem. An evaluation of the proposed method is finally conducted both on synthetic and real data in the specific context of hyperspectral image interpretation, unifying two standard analysis techniques, namely unmixing and classification. |
| Tasks | Representation Learning |
| Published | 2019-02-07 |
| URL | https://arxiv.org/abs/1902.02597v4 |
| https://arxiv.org/pdf/1902.02597v4.pdf | |
| PWC | https://paperswithcode.com/paper/matrix-cofactorization-for-joint |
| Repo | |
| Framework | |
Differentiable probabilistic models of scientific imaging with the Fourier slice theorem
| Title | Differentiable probabilistic models of scientific imaging with the Fourier slice theorem |
| Authors | Karen Ullrich, Rianne van den Berg, Marcus Brubaker, David Fleet, Max Welling |
| Abstract | Scientific imaging techniques such as optical and electron microscopy and computed tomography (CT) scanning are used to study the 3D structure of an object through 2D observations. These observations are related to the original 3D object through orthogonal integral projections. For common 3D reconstruction algorithms, computational efficiency requires the modeling of the 3D structures to take place in Fourier space by applying the Fourier slice theorem. At present, it is unclear how to differentiate through the projection operator, and hence current learning algorithms can not rely on gradient based methods to optimize 3D structure models. In this paper we show how back-propagation through the projection operator in Fourier space can be achieved. We demonstrate the validity of the approach with experiments on 3D reconstruction of proteins. We further extend our approach to learning probabilistic models of 3D objects. This allows us to predict regions of low sampling rates or estimate noise. A higher sample efficiency can be reached by utilizing the learned uncertainties of the 3D structure as an unsupervised estimate of the model fit. Finally, we demonstrate how the reconstruction algorithm can be extended with an amortized inference scheme on unknown attributes such as object pose. Through empirical studies we show that joint inference of the 3D structure and the object pose becomes more difficult when the ground truth object contains more symmetries. Due to the presence of for instance (approximate) rotational symmetries, the pose estimation can easily get stuck in local optima, inhibiting a fine-grained high-quality estimate of the 3D structure. |
| Tasks | 3D Reconstruction, Computed Tomography (CT), Pose Estimation |
| Published | 2019-06-18 |
| URL | https://arxiv.org/abs/1906.07582v2 |
| https://arxiv.org/pdf/1906.07582v2.pdf | |
| PWC | https://paperswithcode.com/paper/differentiable-probabilistic-models-of |
| Repo | |
| Framework | |
Federated Learning Of Out-Of-Vocabulary Words
| Title | Federated Learning Of Out-Of-Vocabulary Words |
| Authors | Mingqing Chen, Rajiv Mathews, Tom Ouyang, Françoise Beaufays |
| Abstract | We demonstrate that a character-level recurrent neural network is able to learn out-of-vocabulary (OOV) words under federated learning settings, for the purpose of expanding the vocabulary of a virtual keyboard for smartphones without exporting sensitive text to servers. High-frequency words can be sampled from the trained generative model by drawing from the joint posterior directly. We study the feasibility of the approach in two settings: (1) using simulated federated learning on a publicly available non-IID per-user dataset from a popular social networking website, (2) using federated learning on data hosted on user mobile devices. The model achieves good recall and precision compared to ground-truth OOV words in setting (1). With (2) we demonstrate the practicality of this approach by showing that we can learn meaningful OOV words with good character-level prediction accuracy and cross entropy loss. |
| Tasks | |
| Published | 2019-03-26 |
| URL | http://arxiv.org/abs/1903.10635v1 |
| http://arxiv.org/pdf/1903.10635v1.pdf | |
| PWC | https://paperswithcode.com/paper/federated-learning-of-out-of-vocabulary-words |
| Repo | |
| Framework | |
A probabilistic assessment of the Indo-Aryan Inner-Outer Hypothesis
| Title | A probabilistic assessment of the Indo-Aryan Inner-Outer Hypothesis |
| Authors | Chundra A. Cathcart |
| Abstract | This paper uses a novel data-driven probabilistic approach to address the century-old Inner-Outer hypothesis of Indo-Aryan. I develop a Bayesian hierarchical mixed-membership model to assess the validity of this hypothesis using a large data set of automatically extracted sound changes operating between Old Indo-Aryan and Modern Indo-Aryan speech varieties. I employ different prior distributions in order to model sound change, one of which, the logistic normal distribution, has not received much attention in linguistics outside of Natural Language Processing, despite its many attractive features. I find evidence for cohesive dialect groups that have made their imprint on contemporary Indo-Aryan languages, and find that when a logistic normal prior is used, the distribution of dialect components across languages is largely compatible with a core-periphery pattern similar to that proposed under the Inner-Outer hypothesis. |
| Tasks | |
| Published | 2019-11-29 |
| URL | https://arxiv.org/abs/1912.01957v1 |
| https://arxiv.org/pdf/1912.01957v1.pdf | |
| PWC | https://paperswithcode.com/paper/191201957 |
| Repo | |
| Framework | |
Dynamic Pricing on E-commerce Platform with Deep Reinforcement Learning
| Title | Dynamic Pricing on E-commerce Platform with Deep Reinforcement Learning |
| Authors | Jiaxi Liu, Yidong Zhang, Xiaoqing Wang, Yuming Deng, Xingyu Wu |
| Abstract | In this paper we present an end-to-end framework for addressing the problem of dynamic pricing on E-commerce platform using methods based on deep reinforcement learning (DRL). By using four groups of different business data to represent the states of each time period, we model the dynamic pricing problem as a Markov Decision Process (MDP). Compared with the state-of-the-art DRL-based dynamic pricing algorithms, our approaches make the following three contributions. First, we extend the discrete set problem to the continuous price set. Second, instead of using revenue as the reward function directly, we define a new function named difference of revenue conversion rates (DRCR). Third, the cold-start problem of MDP is tackled by pre-training and evaluation using some carefully chosen historical sales data. Our approaches are evaluated by both offline evaluation method using real dataset of Alibaba Inc., and online field experiments on Tmall.com, a major online shopping website owned by Alibaba Inc.. In particular, experiment results suggest that DRCR is a more appropriate reward function than revenue, which is widely used by current literature. In the end, field experiments, which last for months on 1000 stock keeping units (SKUs) of products demonstrate that continuous price sets have better performance than discrete sets and show that our approaches significantly outperformed the manual pricing by operation experts. |
| Tasks | |
| Published | 2019-12-05 |
| URL | https://arxiv.org/abs/1912.02572v1 |
| https://arxiv.org/pdf/1912.02572v1.pdf | |
| PWC | https://paperswithcode.com/paper/dynamic-pricing-on-e-commerce-platform-with |
| Repo | |
| Framework | |
Active learning for level set estimation under cost-dependent input uncertainty
| Title | Active learning for level set estimation under cost-dependent input uncertainty |
| Authors | Yu Inatsu, Masayuki Karasuyama, Keiichi Inoue, Ichiro Takeuchi |
| Abstract | As part of a quality control process in manufacturing it is often necessary to test whether all parts of a product satisfy a required property, with as few inspections as possible. When multiple inspection apparatuses with different costs and precision exist, it is desirable that testing can be carried out cost-effectively by properly controlling the trade-off between the costs and the precision. In this paper, we formulate this as a level set estimation (LSE) problem under cost-dependent input uncertainty - LSE being a type of active learning for estimating the level set, i.e., the subset of the input space in which an unknown function value is greater or smaller than a pre-determined threshold. Then, we propose a new algorithm for LSE under cost-dependent input uncertainty with theoretical convergence guarantee. We demonstrate the effectiveness of the proposed algorithm by applying it to synthetic and real datasets. |
| Tasks | Active Learning |
| Published | 2019-09-13 |
| URL | https://arxiv.org/abs/1909.06064v1 |
| https://arxiv.org/pdf/1909.06064v1.pdf | |
| PWC | https://paperswithcode.com/paper/active-learning-for-level-set-estimation |
| Repo | |
| Framework | |
Knowledge-based Biomedical Data Science 2019
| Title | Knowledge-based Biomedical Data Science 2019 |
| Authors | Tiffany J. Callahan, Harrison Pielke-Lombardo, Ignacio J. Tripodi, Lawrence E. Hunter |
| Abstract | Knowledge-based biomedical data science (KBDS) involves the design and implementation of computer systems that act as if they knew about biomedicine. Such systems depend on formally represented knowledge in computer systems, often in the form of knowledge graphs. Here we survey the progress in the last year in systems that use formally represented knowledge to address data science problems in both clinical and biological domains, as well as on approaches for creating knowledge graphs. Major themes include the relationships between knowledge graphs and machine learning, the use of natural language processing, and the expansion of knowledge-based approaches to novel domains, such as Chinese Traditional Medicine and biodiversity. |
| Tasks | Knowledge Graphs |
| Published | 2019-10-08 |
| URL | https://arxiv.org/abs/1910.06710v1 |
| https://arxiv.org/pdf/1910.06710v1.pdf | |
| PWC | https://paperswithcode.com/paper/knowledge-based-biomedical-data-science-2019 |
| Repo | |
| Framework | |
Layer Pruning for Accelerating Very Deep Neural Networks
| Title | Layer Pruning for Accelerating Very Deep Neural Networks |
| Authors | Weiwei Zhang, Changsheng chen, Xuechun Wu, Jialin Gao, Di Bao, Jiwei Li, Xi Zhou |
| Abstract | In this paper, we propose an adaptive pruning method. This method can cut off the channel and layer adaptively. The proportion of the layer and the channel to be cut is learned adaptively. The pruning method proposed in this paper can reduce half of the parameters, and the accuracy will not decrease or even be higher than baseline. |
| Tasks | |
| Published | 2019-10-28 |
| URL | https://arxiv.org/abs/1910.12727v1 |
| https://arxiv.org/pdf/1910.12727v1.pdf | |
| PWC | https://paperswithcode.com/paper/layer-pruning-for-accelerating-very-deep |
| Repo | |
| Framework | |
Prediction of neural network performance by phenotypic modeling
| Title | Prediction of neural network performance by phenotypic modeling |
| Authors | Alexander Hagg, Martin Zaefferer, Jörg Stork, Adam Gaier |
| Abstract | Surrogate models are used to reduce the burden of expensive-to-evaluate objective functions in optimization. By creating models which map genomes to objective values, these models can estimate the performance of unknown inputs, and so be used in place of expensive objective functions. Evolutionary techniques such as genetic programming or neuroevolution commonly alter the structure of the genome itself. A lack of consistency in the genotype is a fatal blow to data-driven modeling techniques: interpolation between points is impossible without a common input space. However, while the dimensionality of genotypes may differ across individuals, in many domains, such as controllers or classifiers, the dimensionality of the input and output remains constant. In this work we leverage this insight to embed differing neural networks into the same input space. To judge the difference between the behavior of two neural networks, we give them both the same input sequence, and examine the difference in output. This difference, the phenotypic distance, can then be used to situate these networks into a common input space, allowing us to produce surrogate models which can predict the performance of neural networks regardless of topology. In a robotic navigation task, we show that models trained using this phenotypic embedding perform as well or better as those trained on the weight values of a fixed topology neural network. We establish such phenotypic surrogate models as a promising and flexible approach which enables surrogate modeling even for representations that undergo structural changes. |
| Tasks | |
| Published | 2019-07-16 |
| URL | https://arxiv.org/abs/1907.07075v1 |
| https://arxiv.org/pdf/1907.07075v1.pdf | |
| PWC | https://paperswithcode.com/paper/prediction-of-neural-network-performance-by |
| Repo | |
| Framework | |
Joint Optimization of Tree-based Index and Deep Model for Recommender Systems
| Title | Joint Optimization of Tree-based Index and Deep Model for Recommender Systems |
| Authors | Han Zhu, Daqing Chang, Ziru Xu, Pengye Zhang, Xiang Li, Jie He, Han Li, Jian Xu, Kun Gai |
| Abstract | Large-scale industrial recommender systems are usually confronted with computational problems due to the enormous corpus size. To retrieve and recommend the most relevant items to users under response time limits, resorting to an efficient index structure is an effective and practical solution. The previous work Tree-based Deep Model (TDM) \cite{zhu2018learning} greatly improves recommendation accuracy using tree index. By indexing items in a tree hierarchy and training a user-node preference prediction model satisfying a max-heap like property in the tree, TDM provides logarithmic computational complexity w.r.t. the corpus size, enabling the use of arbitrary advanced models in candidate retrieval and recommendation. In tree-based recommendation methods, the quality of both the tree index and the user-node preference prediction model determines the recommendation accuracy for the most part. We argue that the learning of tree index and preference model has interdependence. Our purpose, in this paper, is to develop a method to jointly learn the index structure and user preference prediction model. In our proposed joint optimization framework, the learning of index and user preference prediction model are carried out under a unified performance measure. Besides, we come up with a novel hierarchical user preference representation utilizing the tree index hierarchy. Experimental evaluations with two large-scale real-world datasets show that the proposed method improves recommendation accuracy significantly. Online A/B test results at a display advertising platform also demonstrate the effectiveness of the proposed method in production environments. |
| Tasks | Recommendation Systems |
| Published | 2019-02-19 |
| URL | https://arxiv.org/abs/1902.07565v2 |
| https://arxiv.org/pdf/1902.07565v2.pdf | |
| PWC | https://paperswithcode.com/paper/joint-optimization-of-tree-based-index-and |
| Repo | |
| Framework | |
Convergence Rates of Posterior Distributions in Markov Decision Process
| Title | Convergence Rates of Posterior Distributions in Markov Decision Process |
| Authors | Zhen Li, Eric Laber |
| Abstract | In this paper, we show the convergence rates of posterior distributions of the model dynamics in a MDP for both episodic and continuous tasks. The theoretical results hold for general state and action space and the parameter space of the dynamics can be infinite dimensional. Moreover, we show the convergence rates of posterior distributions of the mean accumulative reward under a fixed or the optimal policy and of the regret bound. A variant of Thompson sampling algorithm is proposed which provides both posterior convergence rates for the dynamics and the regret-type bound. Then the previous results are extended to Markov games. Finally, we show numerical results with three simulation scenarios and conclude with discussions. |
| Tasks | |
| Published | 2019-07-22 |
| URL | https://arxiv.org/abs/1907.09083v1 |
| https://arxiv.org/pdf/1907.09083v1.pdf | |
| PWC | https://paperswithcode.com/paper/convergence-rates-of-posterior-distributions |
| Repo | |
| Framework | |
Does AlphaGo actually play Go? Concerning the State Space of Artificial Intelligence
| Title | Does AlphaGo actually play Go? Concerning the State Space of Artificial Intelligence |
| Authors | Holger Lyre |
| Abstract | The overarching goal of this paper is to develop a general model of the state space of AI. Given the breathtaking progress in AI research and technologies in recent years, such conceptual work is of substantial theoretical interest. The present AI hype is mainly driven by the triumph of deep learning neural networks. As the distinguishing feature of such networks is the ability to self-learn, self-learning is identified as one important dimension of the AI state space. Another main dimension lies in the possibility to go over from specific to more general types of problems. The third main dimension is provided by semantic grounding. Since this is a philosophically complex and controversial dimension, a larger part of the paper is devoted to it. We take a fresh look at known foundational arguments in the philosophy of mind and cognition that are gaining new relevance in view of the recent AI developments including the blockhead objection, the Turing test, the symbol grounding problem, the Chinese room argument, and general use-theoretic considerations of meaning. Finally, the AI state space, spanned by the main dimensions generalization, grounding and “selfx-ness”, possessing self-x properties such as self-learning, is outlined. |
| Tasks | |
| Published | 2019-12-13 |
| URL | https://arxiv.org/abs/1912.10005v1 |
| https://arxiv.org/pdf/1912.10005v1.pdf | |
| PWC | https://paperswithcode.com/paper/does-alphago-actually-play-go-concerning-the |
| Repo | |
| Framework | |
A Differential Approach for Gaze Estimation
| Title | A Differential Approach for Gaze Estimation |
| Authors | Gang Liu, Yu Yu, Kenneth A. Funes Mora, Jean-Marc Odobez |
| Abstract | Non-invasive gaze estimation methods usually regress gaze directions directly from a single face or eye image. However, due to important variabilities in eye shapes and inner eye structures amongst individuals, universal models obtain limited accuracies and their output usually exhibit high variance as well as biases which are subject dependent. Therefore, increasing accuracy is usually done through calibration, allowing gaze predictions for a subject to be mapped to his/her actual gaze. In this paper, we introduce a novel image differential method for gaze estimation. We propose to directly train a differential convolutional neural network to predict the gaze differences between two eye input images of the same subject. Then, given a set of subject specific calibration images, we can use the inferred differences to predict the gaze direction of a novel eye sample. The assumption is that by allowing the comparison between two eye images, annoyance factors (alignment, eyelid closing, illumination perturbations) which usually plague single image prediction methods can be much reduced, allowing better prediction altogether. Experiments on 3 public datasets validate our approach which constantly outperforms state-of-the-art methods even when using only one calibration sample or when the latter methods are followed by subject specific gaze adaptation. |
| Tasks | Calibration, Gaze Estimation |
| Published | 2019-04-20 |
| URL | https://arxiv.org/abs/1904.09459v3 |
| https://arxiv.org/pdf/1904.09459v3.pdf | |
| PWC | https://paperswithcode.com/paper/a-differential-approach-for-gaze-estimation |
| Repo | |
| Framework | |
Neural network integral representations with the ReLU activation function
| Title | Neural network integral representations with the ReLU activation function |
| Authors | Anton Dereventsov, Armenak Petrosyan, Clayton Webster |
| Abstract | We derive a formula for the integral representation of a shallow neural network with the ReLU activation function under the finite $L_1$-norm assumption on the outer weights with respect to Lebesgue measure on the sphere. In the case of univariate target functions, we further provide a closed-form formula for all possible representations. Additionally, in this case, our formula allows one to explicitly solve the least $L_1$-norm neural network representation for a given function. |
| Tasks | |
| Published | 2019-10-07 |
| URL | https://arxiv.org/abs/1910.02743v2 |
| https://arxiv.org/pdf/1910.02743v2.pdf | |
| PWC | https://paperswithcode.com/paper/neural-network-integral-representations-with |
| Repo | |
| Framework | |
