January 31, 2020

2948 words 14 mins read

Paper Group ANR 111

Style Transfer Applied to Face Liveness Detection with User-Centered Models. MultiNet++: Multi-Stream Feature Aggregation and Geometric Loss Strategy for Multi-Task Learning. Pitfalls in the Evaluation of Sentence Embeddings. Report on the First Knowledge Graph Reasoning Challenge 2018 – Toward the eXplainable AI System. Asymptotic Consistency of …

Style Transfer Applied to Face Liveness Detection with User-Centered Models


Title	Style Transfer Applied to Face Liveness Detection with User-Centered Models
Authors	Israel A. Laurensi R., Luciana T. Menon, Manoel Camillo O. Penna N., Alessandro L. Koerich, Alceu S. Britto Jr
Abstract	This paper proposes a face anti-spoofing user-centered model (FAS-UCM). The major difficulty, in this case, is obtaining fraudulent images from all users to train the models. To overcome this problem, the proposed method is divided in three main parts: generation of new spoof images, based on style transfer and spoof image representation models; training of a Convolutional Neural Network (CNN) for liveness detection; evaluation of the live and spoof testing images for each subject. The generalization of the CNN to perform style transfer has shown promising qualitative results. Preliminary results have shown that the proposed method is capable of distinguishing between live and spoof images on the SiW database, with an average classification error rate of 0.22.
Tasks	Face Anti-Spoofing, Style Transfer
Published	2019-07-16
URL	https://arxiv.org/abs/1907.07270v1
PDF	https://arxiv.org/pdf/1907.07270v1.pdf
PWC	https://paperswithcode.com/paper/style-transfer-applied-to-face-liveness
Repo
Framework

MultiNet++: Multi-Stream Feature Aggregation and Geometric Loss Strategy for Multi-Task Learning


Title	MultiNet++: Multi-Stream Feature Aggregation and Geometric Loss Strategy for Multi-Task Learning
Authors	Sumanth Chennupati, Ganesh Sistu, Senthil Yogamani, Samir A Rawashdeh
Abstract	Multi-task learning is commonly used in autonomous driving for solving various visual perception tasks. It offers significant benefits in terms of both performance and computational complexity. Current work on multi-task learning networks focus on processing a single input image and there is no known implementation of multi-task learning handling a sequence of images. In this work, we propose a multi-stream multi-task network to take advantage of using feature representations from preceding frames in a video sequence for joint learning of segmentation, depth, and motion. The weights of the current and previous encoder are shared so that features computed in the previous frame can be leveraged without additional computation. In addition, we propose to use the geometric mean of task losses as a better alternative to the weighted average of task losses. The proposed loss function facilitates better handling of the difference in convergence rates of different tasks. Experimental results on KITTI, Cityscapes and SYNTHIA datasets demonstrate that the proposed strategies outperform various existing multi-task learning solutions.
Tasks	Autonomous Driving, Multi-Task Learning
Published	2019-04-15
URL	http://arxiv.org/abs/1904.08492v2
PDF	http://arxiv.org/pdf/1904.08492v2.pdf
PWC	https://paperswithcode.com/paper/multinet-multi-stream-feature-aggregation-and
Repo
Framework

Pitfalls in the Evaluation of Sentence Embeddings


Title	Pitfalls in the Evaluation of Sentence Embeddings
Authors	Steffen Eger, Andreas Rücklé, Iryna Gurevych
Abstract	Deep learning models continuously break new records across different NLP tasks. At the same time, their success exposes weaknesses of model evaluation. Here, we compile several key pitfalls of evaluation of sentence embeddings, a currently very popular NLP paradigm. These pitfalls include the comparison of embeddings of different sizes, normalization of embeddings, and the low (and diverging) correlations between transfer and probing tasks. Our motivation is to challenge the current evaluation of sentence embeddings and to provide an easy-to-access reference for future research. Based on our insights, we also recommend better practices for better future evaluations of sentence embeddings.
Tasks	Sentence Embeddings
Published	2019-06-04
URL	https://arxiv.org/abs/1906.01575v1
PDF	https://arxiv.org/pdf/1906.01575v1.pdf
PWC	https://paperswithcode.com/paper/pitfalls-in-the-evaluation-of-sentence
Repo
Framework

Report on the First Knowledge Graph Reasoning Challenge 2018 – Toward the eXplainable AI System


Title	Report on the First Knowledge Graph Reasoning Challenge 2018 – Toward the eXplainable AI System
Authors	Takahiro Kawamura, Shusaku Egami, Koutarou Tamura, Yasunori Hokazono, Takanori Ugai, Yusuke Koyanagi, Fumihito Nishino, Seiji Okajima, Katsuhiko Murakami, Kunihiko Takamatsu, Aoi Sugiura, Shun Shiramatsu, Shawn Zhang, Kouji Kozaki
Abstract	A new challenge for knowledge graph reasoning started in 2018. Deep learning has promoted the application of artificial intelligence (AI) techniques to a wide variety of social problems. Accordingly, being able to explain the reason for an AI decision is becoming important to ensure the secure and safe use of AI techniques. Thus, we, the Special Interest Group on Semantic Web and Ontology of the Japanese Society for AI, organized a challenge calling for techniques that reason and/or estimate which characters are criminals while providing a reasonable explanation based on an open knowledge graph of a well-known Sherlock Holmes mystery story. This paper presents a summary report of the first challenge held in 2018, including the knowledge graph construction, the techniques proposed for reasoning and/or estimation, the evaluation metrics, and the results. The first prize went to an approach that formalized the problem as a constraint satisfaction problem and solved it using a lightweight formal method; the second prize went to an approach that used SPARQL and rules; the best resource prize went to a submission that constructed word embedding of characters from all sentences of Sherlock Holmes novels; and the best idea prize went to a discussion multi-agents model. We conclude this paper with the plans and issues for the next challenge in 2019.
Tasks	graph construction
Published	2019-08-22
URL	https://arxiv.org/abs/1908.08184v1
PDF	https://arxiv.org/pdf/1908.08184v1.pdf
PWC	https://paperswithcode.com/paper/report-on-the-first-knowledge-graph-reasoning
Repo
Framework

Asymptotic Consistency of Loss-Calibrated Variational Bayes


Title	Asymptotic Consistency of Loss-Calibrated Variational Bayes
Authors	Prateek Jaiswal, Harsha Honnappa, Vinayak A. Rao
Abstract	This paper establishes the asymptotic consistency of the {\it loss-calibrated variational Bayes} (LCVB) method. LCVB was proposed in~\cite{LaSiGh2011} as a method for approximately computing Bayesian posteriors in a `loss aware' manner. This methodology is also highly relevant in general data-driven decision-making contexts. Here, we not only establish the asymptotic consistency of the calibrated approximate posterior, but also the asymptotic consistency of decision rules. We also establish the asymptotic consistency of decision rules obtained from a` naive’ variational Bayesian procedure.
Tasks	Decision Making
Published	2019-11-04
URL	https://arxiv.org/abs/1911.01288v1
PDF	https://arxiv.org/pdf/1911.01288v1.pdf
PWC	https://paperswithcode.com/paper/asymptotic-consistency-of-loss-calibrated
Repo
Framework

Tomographic Reconstruction of Triaxial Strain Fields from Bragg-Edge Neutron Imaging


Title	Tomographic Reconstruction of Triaxial Strain Fields from Bragg-Edge Neutron Imaging
Authors	J. N. Hendriks, A. W. T. Gregg, R. R. Jackson, C. M. Wensrich, A. Wills, A. S. Tremsin, T. Shinohara, V. Luzin, O. Kirstein
Abstract	This paper presents a proof-of-concept demonstration of triaxial strain tomography from Bragg-edge neutron imaging within a three-dimensional sample. Bragg-edge neutron transmission can provide high-resolution images of the average through thickness strain within a polycrystalline material. This poses an associated rich tomography problem which seeks to reconstruct the full triaxial strain field from these images. The presented demonstration is an important step towards solving this problem, and towards a technique capable of studying the residual strain and stress within engineering components. A Gaussian process based approach is used that ensures the reconstruction satisfies equilibrium and known boundary conditions. This approach is demonstrated experimentally on a non-trivial steel sample with use of the RADEN instrument at the Japan Proton Accelerator Research Complex. Validation of the reconstruction is provided by comparison with conventional strain scans from the KOWARI constant-wavelength strain diffractometer at the Australian Nuclear Science and Technology Organisation and simulations via finite element analysis.
Tasks
Published	2019-06-20
URL	https://arxiv.org/abs/1906.08506v4
PDF	https://arxiv.org/pdf/1906.08506v4.pdf
PWC	https://paperswithcode.com/paper/tomographic-reconstruction-of-triaxial-strain
Repo
Framework

Real-Time Panoptic Segmentation from Dense Detections


Title	Real-Time Panoptic Segmentation from Dense Detections
Authors	Rui Hou, Jie Li, Arjun Bhargava, Allan Raventos, Vitor Guizilini, Chao Fang, Jerome Lynch, Adrien Gaidon
Abstract	Panoptic segmentation is a complex full scene parsing task requiring simultaneous instance and semantic segmentation at high resolution. Current state-of-the-art approaches cannot run in real-time, and simplifying these architectures to improve efficiency severely degrades their accuracy. In this paper, we propose a new single-shot panoptic segmentation network that leverages dense detections and a global self-attention mechanism to operate in real-time with performance approaching the state of the art. We introduce a novel parameter-free mask construction method that substantially reduces computational complexity by efficiently reusing information from the object detection and semantic segmentation sub-tasks. The resulting network has a simple data flow that does not require feature map re-sampling or clustering post-processing, enabling significant hardware acceleration. Our experiments on the Cityscapes and COCO benchmarks show that our network works at 30 FPS on 1024x2048 resolution, trading a 3% relative performance degradation from the current state of the art for up to 440% faster inference.
Tasks	Object Detection, Panoptic Segmentation, Scene Parsing, Semantic Segmentation
Published	2019-12-03
URL	https://arxiv.org/abs/1912.01202v2
PDF	https://arxiv.org/pdf/1912.01202v2.pdf
PWC	https://paperswithcode.com/paper/real-time-panoptic-segmentation-from-dense
Repo
Framework

On the well-posedness of uncalibrated photometric stereo under general lighting


Title	On the well-posedness of uncalibrated photometric stereo under general lighting
Authors	Mohammed Brahimi, Yvain Quéau, Bjoern Haefner, Daniel Cremers
Abstract	Uncalibrated photometric stereo aims at estimating the 3D-shape of a surface, given a set of images captured from the same viewing angle, but under unknown, varying illumination. While the theoretical foundations of this inverse problem under directional lighting are well-established, there is a lack of mathematical evidence for the uniqueness of a solution under general lighting. On the other hand, stable and accurate heuristical solutions of uncalibrated photometric stereo under such general lighting have recently been proposed. The quality of the results demonstrated therein tends to indicate that the problem may actually be well-posed, but this still has to be established. The present paper addresses this theoretical issue, considering first-order spherical harmonics approximation of general lighting. Two important theoretical results are established. First, the orthographic integrability constraint ensures uniqueness of a solution up to a global concave-convex ambiguity, which had already been conjectured, yet not proven. Second, the perspective integrability constraint makes the problem well-posed, which generalizes a previous result limited to directional lighting. Eventually, a closed-form expression for the unique least-squares solution of the problem under perspective projection is provided, allowing numerical simulations on synthetic data to empirically validate our findings.
Tasks
Published	2019-11-17
URL	https://arxiv.org/abs/1911.07268v1
PDF	https://arxiv.org/pdf/1911.07268v1.pdf
PWC	https://paperswithcode.com/paper/on-the-well-posedness-of-uncalibrated
Repo
Framework

Characterizing and Predicting Repeat Food Consumption Behavior for Just-in-Time Interventions


Title	Characterizing and Predicting Repeat Food Consumption Behavior for Just-in-Time Interventions
Authors	Yue Liu, Helena Lee, Palakorn Achananuparp, Ee-Peng Lim, Tzu-Ling Cheng, Shou-De Lin
Abstract	Human beings are creatures of habit. In their daily life, people tend to repeatedly consume similar types of food items over several days and occasionally switch to consuming different types of items when the consumptions become overly monotonous. However, the novel and repeat consumption behaviors have not been studied in food recommendation research. More importantly, the ability to predict daily eating habits of individuals is crucial to improve the effectiveness of food recommender systems in facilitating healthy lifestyle change. In this study, we analyze the patterns of repeat food consumptions using large-scale consumption data from a popular online fitness community called MyFitnessPal (MFP), conduct an offline evaluation of various state-of-the-art algorithms in predicting the next-day food consumption, and analyze their performance across different demographic groups and contexts. The experiment results show that algorithms incorporating the exploration-and-exploitation and temporal dynamics are more effective in the next-day recommendation task than most state-of-the-art algorithms.
Tasks	Recommendation Systems
Published	2019-09-17
URL	https://arxiv.org/abs/1909.07683v1
PDF	https://arxiv.org/pdf/1909.07683v1.pdf
PWC	https://paperswithcode.com/paper/characterizing-and-predicting-repeat-food
Repo
Framework

Revisiting hard thresholding for DNN pruning


Title	Revisiting hard thresholding for DNN pruning
Authors	Konstantinos Pitas, Mike Davies, Pierre Vandergheynst
Abstract	The most common method for DNN pruning is hard thresholding of network weights, followed by retraining to recover any lost accuracy. Recently developed smart pruning algorithms use the DNN response over the training set for a variety of cost functions to determine redundant network weights, leading to less accuracy degradation and possibly less retraining time. For experiments on the total pruning time (pruning time + retraining time) we show that hard thresholding followed by retraining remains the most efficient way of reducing the number of network parameters. However smart pruning algorithms still have advantages when retraining is not possible. In this context we propose a novel smart pruning algorithm based on difference of convex functions optimisation and show that it is often orders of magnitude faster than competing approaches while achieving the lowest classification accuracy degradation. Furthermore we investigate theoretically the effect of hard thresholding on DNN accuracy. We show that accuracy degradation increases with remaining network depth from the pruned layer. We also discover a link between the latent dimensionality of the training data manifold and network robustness to hard thresholding.
Tasks
Published	2019-05-21
URL	https://arxiv.org/abs/1905.08793v1
PDF	https://arxiv.org/pdf/1905.08793v1.pdf
PWC	https://paperswithcode.com/paper/revisiting-hard-thresholding-for-dnn-pruning
Repo
Framework

FQ-Conv: Fully Quantized Convolution for Efficient and Accurate Inference


Title	FQ-Conv: Fully Quantized Convolution for Efficient and Accurate Inference
Authors	Bram-Ernst Verhoef, Nathan Laubeuf, Stefan Cosemans, Peter Debacker, Ioannis Papistas, Arindam Mallik, Diederik Verkest
Abstract	Deep neural networks (DNNs) can be made hardware-efficient by reducing the numerical precision of the weights and activations of the network and by improving the network’s resilience to noise. However, this gain in efficiency often comes at the cost of significantly reduced accuracy. In this paper, we present a novel approach to quantizing convolutional neural network. The resulting networks perform all computations in low-precision, without requiring higher-precision BN and nonlinearities, while still being highly accurate. To achieve this result, we employ a novel quantization technique that learns to optimally quantize the weights and activations of the network during training. Additionally, to enhance training convergence we use a new training technique, called gradual quantization. We leverage the nonlinear and normalizing behavior of our quantization function to effectively remove the higher-precision nonlinearities and BN from the network. The resulting convolutional layers are fully quantized to low precision, from input to output, ideal for neural network accelerators on the edge. We demonstrate the potential of this approach on different datasets and networks, showing that ternary-weight CNNs with low-precision in- and outputs perform virtually on par with their full-precision equivalents. Finally, we analyze the influence of noise on the weights, activations and convolution outputs (multiply-accumulate, MAC) and propose a strategy to improve network performance under noisy conditions.
Tasks	Quantization
Published	2019-12-19
URL	https://arxiv.org/abs/1912.09356v1
PDF	https://arxiv.org/pdf/1912.09356v1.pdf
PWC	https://paperswithcode.com/paper/fq-conv-fully-quantized-convolution-for
Repo
Framework

A Queuing Approach to Parking: Modeling, Verification, and Prediction


Title	A Queuing Approach to Parking: Modeling, Verification, and Prediction
Authors	Hamidreza Tavafoghi, Kameshwar Poolla, Pravin Varaiya
Abstract	We present a queuing model of parking dynamics and a model-based prediction method to provide real-time probabilistic forecasts of future parking occupancy. The queuing model has a non-homogeneous arrival rate and time-varying service time distribution. All statistical assumptions of the model are verified using data from 29 truck parking locations, each with between 55 and 299 parking spots. For each location and each spot the data specifies the arrival and departure times of a truck, for 16 months of operation. The modeling framework presented in this paper provides empirical support for queuing models adopted in many theoretical studies and policy designs. We discuss how our framework can be used to study parking problems in different environments. Based on the queuing model, we propose two prediction methods, a microscopic method and a macroscopic method, that provide a real-time probabilistic forecast of parking occupancy for an arbitrary forecast horizon. These model-based methods convert a probabilistic forecast problem into a parameter estimation problem that can be tackled using classical estimation methods such as regressions or pure machine learning algorithms. We characterize a lower bound for an arbitrary real-time prediction algorithm. We evaluate the performance of these methods using the truck data comparing the outcomes of their implementations with other model-based and model-free methods proposed in the literature.
Tasks
Published	2019-08-29
URL	https://arxiv.org/abs/1908.11479v1
PDF	https://arxiv.org/pdf/1908.11479v1.pdf
PWC	https://paperswithcode.com/paper/a-queuing-approach-to-parking-modeling
Repo
Framework

Towards an Adaptive Robot for Sports and Rehabilitation Coaching


Title	Towards an Adaptive Robot for Sports and Rehabilitation Coaching
Authors	Martin K. Ross, Frank Broz, Lynne Baillie
Abstract	The work presented in this paper aims to explore how, and to what extent, an adaptive robotic coach has the potential to provide extra motivation to adhere to long-term rehabilitation and help fill the coaching gap which occurs during repetitive solo practice in high performance sport. Adapting the behavior of a social robot to a specific user, using reinforcement learning (RL), could be a way of increasing adherence to an exercise routine in both domains. The requirements gathering phase is underway and is presented in this paper along with the rationale of using RL in this context.
Tasks
Published	2019-09-13
URL	https://arxiv.org/abs/1909.08052v1
PDF	https://arxiv.org/pdf/1909.08052v1.pdf
PWC	https://paperswithcode.com/paper/towards-an-adaptive-robot-for-sports-and
Repo
Framework

Weakly-Supervised Deep Learning for Domain Invariant Sentiment Classification


Title	Weakly-Supervised Deep Learning for Domain Invariant Sentiment Classification
Authors	Pratik Kayal, Mayank Singh, Pawan Goyal
Abstract	The task of learning a sentiment classification model that adapts well to any target domain, different from the source domain, is a challenging problem. Majority of the existing approaches focus on learning a common representation by leveraging both source and target data during training. In this paper, we introduce a two-stage training procedure that leverages weakly supervised datasets for developing simple lift-and-shift-based predictive models without being exposed to the target domain during the training phase. Experimental results show that transfer with weak supervision from a source domain to various target domains provides performance very close to that obtained via supervised training on the target domain itself.
Tasks	Sentiment Analysis
Published	2019-10-29
URL	https://arxiv.org/abs/1910.13425v2
PDF	https://arxiv.org/pdf/1910.13425v2.pdf
PWC	https://paperswithcode.com/paper/191013425
Repo
Framework

Semantically Constrained Multilayer Annotation: The Case of Coreference


Title	Semantically Constrained Multilayer Annotation: The Case of Coreference
Authors	Jakob Prange, Nathan Schneider, Omri Abend
Abstract	We propose a coreference annotation scheme as a layer on top of the Universal Conceptual Cognitive Annotation foundational layer, treating units in predicate-argument structure as a basis for entity and event mentions. We argue that this allows coreference annotators to sidestep some of the challenges faced in other schemes, which do not enforce consistency with predicate-argument structure and vary widely in what kinds of mentions they annotate and how. The proposed approach is examined with a pilot annotation study and compared with annotations from other schemes.
Tasks
Published	2019-06-03
URL	https://arxiv.org/abs/1906.00663v3
PDF	https://arxiv.org/pdf/1906.00663v3.pdf
PWC	https://paperswithcode.com/paper/190600663
Repo
Framework