January 26, 2020

2879 words 14 mins read

Paper Group ANR 1454

Information-Theoretic Confidence Bounds for Reinforcement Learning. A General $\mathcal{O}(n^2)$ Hyper-Parameter Optimization for Gaussian Process Regression with Cross-Validation and Non-linearly Constrained ADMM. OneGAN: Simultaneous Unsupervised Learning of Conditional Image Generation, Foreground Segmentation, and Fine-Grained Clustering. Towar …

Information-Theoretic Confidence Bounds for Reinforcement Learning


Title	Information-Theoretic Confidence Bounds for Reinforcement Learning
Authors	Xiuyuan Lu, Benjamin Van Roy
Abstract	We integrate information-theoretic concepts into the design and analysis of optimistic algorithms and Thompson sampling. By making a connection between information-theoretic quantities and confidence bounds, we obtain results that relate the per-period performance of the agent with its information gain about the environment, thus explicitly characterizing the exploration-exploitation tradeoff. The resulting cumulative regret bound depends on the agent’s uncertainty over the environment and quantifies the value of prior information. We show applicability of this approach to several environments, including linear bandits, tabular MDPs, and factored MDPs. These examples demonstrate the potential of a general information-theoretic approach for the design and analysis of reinforcement learning algorithms.
Tasks
Published	2019-11-21
URL	https://arxiv.org/abs/1911.09724v1
PDF	https://arxiv.org/pdf/1911.09724v1.pdf
PWC	https://paperswithcode.com/paper/information-theoretic-confidence-bounds-for-1
Repo
Framework

A General $\mathcal{O}(n^2)$ Hyper-Parameter Optimization for Gaussian Process Regression with Cross-Validation and Non-linearly Constrained ADMM


Title	A General $\mathcal{O}(n^2)$ Hyper-Parameter Optimization for Gaussian Process Regression with Cross-Validation and Non-linearly Constrained ADMM
Authors	Linning Xu, Feng Yin, Jiawei Zhang, Zhi-Quan Luo, Shuguang Cui
Abstract	Hyper-parameter optimization remains as the core issue of Gaussian process (GP) for machine learning nowadays. The benchmark method using maximum likelihood (ML) estimation and gradient descent (GD) is impractical for processing big data due to its $O(n^3)$ complexity. Many sophisticated global or local approximation models, for instance, sparse GP, distributed GP, have been proposed to address such complexity issue. In this paper, we propose two novel and general-purpose GP hyper-parameter training schemes (GPCV-ADMM) by replacing ML with cross-validation (CV) as the fitting criterion and replacing GD with a non-linearly constrained alternating direction method of multipliers (ADMM) as the optimization method. The proposed schemes are of $O(n^2)$ complexity for any covariance matrix without special structure. We conduct various experiments based on both synthetic and real data sets, wherein the proposed schemes show excellent performance in terms of convergence, hyper-parameter estimation accuracy, and computational time in comparison with the traditional ML based routines given in the GPML toolbox.
Tasks
Published	2019-06-06
URL	https://arxiv.org/abs/1906.02387v2
PDF	https://arxiv.org/pdf/1906.02387v2.pdf
PWC	https://paperswithcode.com/paper/a-general-mathcalon2-hyper-parameter
Repo
Framework

OneGAN: Simultaneous Unsupervised Learning of Conditional Image Generation, Foreground Segmentation, and Fine-Grained Clustering


Title	OneGAN: Simultaneous Unsupervised Learning of Conditional Image Generation, Foreground Segmentation, and Fine-Grained Clustering
Authors	Yaniv Benny, Lior Wolf
Abstract	We present a method for simultaneously learning, in an unsupervised manner, (i) a conditional image generator, (ii) foreground extraction and segmentation, (iii) clustering into a two-level class hierarchy, and (iv) object removal and background completion, all done without any use of annotation. The method combines a generative adversarial network and a variational autoencoder, with multiple encoders, generators and discriminators, and benefits from solving all tasks at once. The input to the training scheme is a varied collection of unlabeled images from the same domain, as well as a set of background images without a foreground object. In addition, the image generator can mix the background from one image, with a foreground that is conditioned either on that of a second image or on the index of a desired cluster. The method obtains state of the art results in comparison to the literature methods, when compared to the current state of the art in each of the tasks.
Tasks	Conditional Image Generation, Image Generation
Published	2019-12-31
URL	https://arxiv.org/abs/1912.13471v1
PDF	https://arxiv.org/pdf/1912.13471v1.pdf
PWC	https://paperswithcode.com/paper/onegan-simultaneous-unsupervised-learning-of
Repo
Framework

Towards Domain Adaptation from Limited Data for Question Answering Using Deep Neural Networks


Title	Towards Domain Adaptation from Limited Data for Question Answering Using Deep Neural Networks
Authors	Timothy J. Hazen, Shehzaad Dhuliawala, Daniel Boies
Abstract	This paper explores domain adaptation for enabling question answering (QA) systems to answer questions posed against documents in new specialized domains. Current QA systems using deep neural network (DNN) technology have proven effective for answering general purpose factoid-style questions. However, current general purpose DNN models tend to be ineffective for use in new specialized domains. This paper explores the effectiveness of transfer learning techniques for this problem. In experiments on question answering in the automobile manual domain we demonstrate that standard DNN transfer learning techniques work surprisingly well in adapting DNN models to a new domain using limited amounts of annotated training data in the new domain.
Tasks	Domain Adaptation, Question Answering, Transfer Learning
Published	2019-11-06
URL	https://arxiv.org/abs/1911.02655v1
PDF	https://arxiv.org/pdf/1911.02655v1.pdf
PWC	https://paperswithcode.com/paper/towards-domain-adaptation-from-limited-data
Repo
Framework

Lane Change Decision-making through Deep Reinforcement Learning with Rule-based Constraints


Title	Lane Change Decision-making through Deep Reinforcement Learning with Rule-based Constraints
Authors	Junjie Wang, Qichao Zhang, Dongbin Zhao, Yaran Chen
Abstract	Autonomous driving decision-making is a great challenge due to the complexity and uncertainty of the traffic environment. Combined with the rule-based constraints, a Deep Q-Network (DQN) based method is applied for autonomous driving lane change decision-making task in this study. Through the combination of high-level lateral decision-making and low-level rule-based trajectory modification, a safe and efficient lane change behavior can be achieved. With the setting of our state representation and reward function, the trained agent is able to take appropriate actions in a real-world-like simulator. The generated policy is evaluated on the simulator for 10 times, and the results demonstrate that the proposed rule-based DQN method outperforms the rule-based approach and the DQN method.
Tasks	Autonomous Driving, Decision Making
Published	2019-03-30
URL	http://arxiv.org/abs/1904.00231v2
PDF	http://arxiv.org/pdf/1904.00231v2.pdf
PWC	https://paperswithcode.com/paper/lane-change-decision-making-through-deep
Repo
Framework

Semantic denoising autoencoders for retinal optical coherence tomography


Title	Semantic denoising autoencoders for retinal optical coherence tomography
Authors	Max-Heinrich Laves, Sontje Ihler, Lüder Alexander Kahrs, Tobias Ortmaier
Abstract	Noise in speckle-prone optical coherence tomography tends to obfuscate important details necessary for medical diagnosis. In this paper, a denoising approach that preserves disease characteristics on retinal optical coherence tomography images in ophthalmology is presented. By combining a deep convolutional autoencoder with a priorly trained ResNet image classifier as regularizer, the perceptibility of delicate details is encouraged and only information-less background noise is filtered out. With our approach, higher peak signal-to-noise ratios with $ \mathrm{PSNR} = 31.2,\mathrm{dB} $ and higher classification accuracy of $\mathrm{ACC} = 85.0,%$ can be achieved for denoised images compared to state-of-the-art denoising with $ \mathrm{PSNR} = 29.4,\mathrm{dB} $ or $\mathrm{ACC} = 70.3,%$, depending on the method. It is shown that regularized autoencoders are capable of denoising retinal OCT images without blurring details of diseases.
Tasks	Denoising, Medical Diagnosis
Published	2019-03-23
URL	http://arxiv.org/abs/1903.09809v1
PDF	http://arxiv.org/pdf/1903.09809v1.pdf
PWC	https://paperswithcode.com/paper/semantic-denoising-autoencoders-for-retinal
Repo
Framework

Towards Hierarchical Importance Attribution: Explaining Compositional Semantics for Neural Sequence Models


Title	Towards Hierarchical Importance Attribution: Explaining Compositional Semantics for Neural Sequence Models
Authors	Xisen Jin, Junyi Du, Zhongyu Wei, Xiangyang Xue, Xiang Ren
Abstract	The impressive performance of neural networks on natural language processing tasks attributes to their ability to model complicated word and phrase interactions. Existing flat, word level explanations of predictions hardly unveil how neural networks handle compositional semantics to reach predictions. To tackle the challenge, we study hierarchical explanation of neural network predictions. We identify non-additivity and independent importance attributions within hierarchies as two desirable properties for highlighting word and phrase interactions. We show prior efforts on hierarchical explanations, e.g. contextual decomposition, however, do not satisfy the desired properties mathematically. In this paper, we propose a formal way to quantify the importance of each word or phrase for hierarchical explanations. Following the formulation, we propose Sampling and Contextual Decomposition (SCD) algorithm and Sampling and Occlusion (SOC) algorithm. Human and metrics evaluation on both LSTM models and BERT Transformer models on multiple datasets show that our algorithms outperform prior hierarchical explanation algorithms. Our algorithms apply to hierarchical visualization of compositional semantics, extraction of classification rules and improving human trust of models.
Tasks
Published	2019-11-08
URL	https://arxiv.org/abs/1911.06194v1
PDF	https://arxiv.org/pdf/1911.06194v1.pdf
PWC	https://paperswithcode.com/paper/towards-hierarchical-importance-attribution-1
Repo
Framework

Exposure Interpolation Via Fusing Conventional and Deep Learning Methods


Title	Exposure Interpolation Via Fusing Conventional and Deep Learning Methods
Authors	Chaobing Zheng, Zhengguo Li, Yi Yang, Shiqian Wu
Abstract	Deep learning based methods have penetrated many image processing problems and become dominant solutions to these problems. A natural question raised here is “Is there any space for conventional methods on these problems?” In this paper, exposure interpolation is taken as an example to answer this question and the answer is “Yes”. A framework on fusing conventional and deep learning method is introduced to generate an medium exposure image for two large-exposureratio images. Experimental results indicate that the quality of the medium exposure image is increased significantly through using the deep learning method to refine the interpolated image via the conventional method. The conventional method can be adopted to improve the convergence speed of the deep learning method and to reduce the number of samples which is required by the deep learning method.
Tasks
Published	2019-05-09
URL	https://arxiv.org/abs/1905.03890v4
PDF	https://arxiv.org/pdf/1905.03890v4.pdf
PWC	https://paperswithcode.com/paper/exposure-interpolation-via-fusing
Repo
Framework

Pyramid Vector Quantization and Bit Level Sparsity in Weights for Efficient Neural Networks Inference


Title	Pyramid Vector Quantization and Bit Level Sparsity in Weights for Efficient Neural Networks Inference
Authors	Vincenzo Liguori
Abstract	This paper discusses three basic blocks for the inference of convolutional neural networks (CNNs). Pyramid Vector Quantization (PVQ) is discussed as an effective quantizer for CNNs weights resulting in highly sparse and compressible networks. Properties of PVQ are exploited for the elimination of multipliers during inference while maintaining high performance. The result is then extended to any other quantized weights. The Tiny Yolo v3 CNN is used to compare such basic blocks.
Tasks	Quantization
Published	2019-11-24
URL	https://arxiv.org/abs/1911.10636v1
PDF	https://arxiv.org/pdf/1911.10636v1.pdf
PWC	https://paperswithcode.com/paper/pyramid-vector-quantization-and-bit-level
Repo
Framework

AI-CARGO: A Data-Driven Air-Cargo Revenue Management System


Title	AI-CARGO: A Data-Driven Air-Cargo Revenue Management System
Authors	Stefano Giovanni Rizzo, Ji Lucas, Zoi Kaoudi, Jorge-Arnulfo Quiane-Ruiz, Sanjay Chawla
Abstract	We propose AI-CARGO, a revenue management system for air-cargo that combines machine learning prediction with decision-making using mathematical optimization methods. AI-CARGO addresses a problem that is unique to the air-cargo business, namely the wide discrepancy between the quantity (weight or volume) that a shipper will book and the actual received amount at departure time by the airline. The discrepancy results in sub-optimal and inefficient behavior by both the shipper and the airline resulting in the overall loss of potential revenue for the airline. AI-CARGO also includes a data cleaning component to deal with the heterogeneous forms in which booking data is transmitted to the airline cargo system. AI-CARGO is deployed in the production environment of a large commercial airline company. We have validated the benefits of AI-CARGO using real and synthetic datasets. Especially, we have carried out simulations using dynamic programming techniques to elicit the impact on offloading costs and revenue generation of our proposed system. Our results suggest that combining prediction within a decision-making framework can help dramatically to reduce offloading costs and optimize revenue generation.
Tasks	Decision Making
Published	2019-05-22
URL	https://arxiv.org/abs/1905.09130v1
PDF	https://arxiv.org/pdf/1905.09130v1.pdf
PWC	https://paperswithcode.com/paper/ai-cargo-a-data-driven-air-cargo-revenue
Repo
Framework

SELF: Learning to Filter Noisy Labels with Self-Ensembling


Title	SELF: Learning to Filter Noisy Labels with Self-Ensembling
Authors	Duc Tam Nguyen, Chaithanya Kumar Mummadi, Thi Phuong Nhung Ngo, Thi Hoai Phuong Nguyen, Laura Beggel, Thomas Brox
Abstract	Deep neural networks (DNNs) have been shown to over-fit a dataset when being trained with noisy labels for a long enough time. To overcome this problem, we present a simple and effective method self-ensemble label filtering (SELF) to progressively filter out the wrong labels during training. Our method improves the task performance by gradually allowing supervision only from the potentially non-noisy (clean) labels and stops learning on the filtered noisy labels. For the filtering, we form running averages of predictions over the entire training dataset using the network output at different training epochs. We show that these ensemble estimates yield more accurate identification of inconsistent predictions throughout training than the single estimates of the network at the most recent training epoch. While filtered samples are removed entirely from the supervised training loss, we dynamically leverage them via semi-supervised learning in the unsupervised loss. We demonstrate the positive effect of such an approach on various image classification tasks under both symmetric and asymmetric label noise and at different noise ratios. It substantially outperforms all previous works on noise-aware learning across different datasets and can be applied to a broad set of network architectures.
Tasks	Image Classification
Published	2019-10-04
URL	https://arxiv.org/abs/1910.01842v1
PDF	https://arxiv.org/pdf/1910.01842v1.pdf
PWC	https://paperswithcode.com/paper/self-learning-to-filter-noisy-labels-with
Repo
Framework

Forecast Aggregation via Peer Prediction


Title	Forecast Aggregation via Peer Prediction
Authors	Juntao Wang, Yang Liu, Yiling Chen
Abstract	Crowdsourcing is a popular paradigm for soliciting forecasts on future events. As people may have different forecasts, how to aggregate solicited forecasts into a single accurate prediction remains to be an important challenge, especially when no historical accuracy information is available for identifying experts. In this paper, we borrow ideas from the peer prediction literature and assess the prediction accuracy of participants using solely the collected forecasts. This approach leverages the correlations among peer reports to cross-validate each participant’s forecasts and allows us to assign a “peer assessment score (PAS)” for each agent as a proxy for the agent’s prediction accuracy. We identify several empirically effective methods to generate PAS and propose an aggregation framework that uses PAS to identify experts and to boost existing aggregators’ prediction accuracy. We evaluate our methods over 14 real-world datasets and show that i) PAS generated from peer prediction methods can approximately reflect the prediction accuracy of agents, and ii) our aggregation framework demonstrates consistent and significant improvement in the prediction accuracy over existing aggregators for both binary and multi-choice questions under three popular accuracy measures: Brier score (mean square error), log score (cross-entropy loss) and AUC-ROC.
Tasks
Published	2019-10-09
URL	https://arxiv.org/abs/1910.03779v3
PDF	https://arxiv.org/pdf/1910.03779v3.pdf
PWC	https://paperswithcode.com/paper/forecast-aggregation-via-peer-prediction
Repo
Framework

Compressed Sensing Microscopy with Scanning Line Probes


Title	Compressed Sensing Microscopy with Scanning Line Probes
Authors	Han-Wen Kuo, Anna E. Dorfi, Daniel V. Esposito, John N. Wright
Abstract	In applications of scanning probe microscopy, images are acquired by raster scanning a point probe across a sample. Viewed from the perspective of compressed sensing (CS), this pointwise sampling scheme is inefficient, especially when the target image is structured. While replacing point measurements with delocalized, incoherent measurements has the potential to yield order-of-magnitude improvements in scan time, implementing the delocalized measurements of CS theory is challenging. In this paper we study a partially delocalized probe construction, in which the point probe is replaced with a continuous line, creating a sensor which essentially acquires line integrals of the target image. We show through simulations, rudimentary theoretical analysis, and experiments, that these line measurements can image sparse samples far more efficiently than traditional point measurements, provided the local features in the sample are enough separated. Despite this promise, practical reconstruction from line measurements poses additional difficulties: the measurements are partially coherent, and real measurements exhibit nonidealities. We show how to overcome these limitations using natural strategies (reweighting to cope with coherence, blind calibration for nonidealities), culminating in an end-to-end demonstration.
Tasks	Calibration
Published	2019-09-26
URL	https://arxiv.org/abs/1909.12342v1
PDF	https://arxiv.org/pdf/1909.12342v1.pdf
PWC	https://paperswithcode.com/paper/compressed-sensing-microscopy-with-scanning
Repo
Framework

Multiple Patients Behavior Detection in Real-time using mmWave Radar and Deep CNNs


Title	Multiple Patients Behavior Detection in Real-time using mmWave Radar and Deep CNNs
Authors	Feng Jin, Renyuan Zhang, Arindam Sengupta, Siyang Cao, Salim Hariri, Nimit K. Agarwal, Sumit K. Agarwal
Abstract	To address potential gaps noted in patient monitoring in the hospital, a novel patient behavior detection system using mmWave radar and deep convolution neural network (CNN), which supports the simultaneous recognition of multiple patients’ behaviors in real-time, is proposed. In this study, we use an mmWave radar to track multiple patients and detect the scattering point cloud of each one. For each patient, the Doppler pattern of the point cloud over a time period is collected as the behavior signature. A three-layer CNN model is created to classify the behavior for each patient. The tracking and point clouds detection algorithm was also implemented on an mmWave radar hardware platform with an embedded graphics processing unit (GPU) board to collect Doppler pattern and run the CNN model. A training dataset of six types of behavior were collected, over a long duration, to train the model using Adam optimizer with an objective to minimize cross-entropy loss function. Lastly, the system was tested for real-time operation and obtained a very good inference accuracy when predicting each patient’s behavior in a two-patient scenario.
Tasks
Published	2019-11-14
URL	https://arxiv.org/abs/1911.06363v1
PDF	https://arxiv.org/pdf/1911.06363v1.pdf
PWC	https://paperswithcode.com/paper/multiple-patients-behavior-detection-in-real
Repo
Framework

Distance-Preserving Graph Embeddings from Random Neural Features


Title	Distance-Preserving Graph Embeddings from Random Neural Features
Authors	Daniele Zambon, Cesare Alippi, Lorenzo Livi
Abstract	We present Graph Random Neural Features (GRNF), a novel embedding method from graph-structured data to real vectors based on a family of graph neural networks. The embedding naturally deals with graph isomorphism and preserves, in probability, the metric structure of graph domain. In addition to being an explicit embedding method, it also allows to efficiently and effectively approximate graph metric distances (as well as complete kernel functions); a criterion to select the embedding dimension trading off the approximation accuracy with the computational cost is also provided. Derived GRNF can be used within traditional processing methods or as input layer of a larger graph neural network. The theoretical guarantees that accompany GRNF ensure that the considered graph distance is metric, hence allowing to distinguish any pair of non-isomorphic graphs.
Tasks
Published	2019-09-09
URL	https://arxiv.org/abs/1909.03790v2
PDF	https://arxiv.org/pdf/1909.03790v2.pdf
PWC	https://paperswithcode.com/paper/distance-preserving-graph-embeddings-from
Repo
Framework