Paper Group NAWR 5
Learning a Unified Classifier Incrementally via Rebalancing. Partially Encrypted Deep Learning using Functional Encryption. Approximate Bayesian Inference for a Mechanistic Model of Vesicle Release at a Ribbon Synapse. The MuCoW Test Suite at WMT 2019: Automatically Harvested Multilingual Contrastive Word Sense Disambiguation Test Sets for Machine …
Learning a Unified Classifier Incrementally via Rebalancing
| Title | Learning a Unified Classifier Incrementally via Rebalancing |
| Authors | Saihui Hou, Xinyu Pan, Chen Change Loy, Zilei Wang, Dahua Lin |
| Abstract | Conventionally, deep neural networks are trained offline, relying on a large dataset prepared in advance. This paradigm is often challenged in real-world applications, e.g. online services that involve continuous streams of incoming data. Recently, incremental learning receives increasing attention, and is considered as a promising solution to the practical challenges mentioned above. However, it has been observed that incremental learning is subject to a fundamental difficulty – catastrophic forgetting, namely adapting a model to new data often results in severe performance degradation on previous tasks or classes. Our study reveals that the imbalance between previous and new data is a crucial cause to this problem. In this work, we develop a new framework for incrementally learning a unified classifier, e.g. a classifier that treats both old and new classes uniformly. Specifically, we incorporate three components, cosine normalization, less-forget constraint, and inter-class separation, to mitigate the adverse effects of the imbalance. Experiments show that the proposed method can effectively rebalance the training process, thus obtaining superior performance compared to the existing methods. On CIFAR-100 and ImageNet, our method can reduce the classification errors by more than 6% and 13% respectively, under the incremental setting of 10 phases. |
| Tasks | |
| Published | 2019-06-01 |
| URL | http://openaccess.thecvf.com/content_CVPR_2019/html/Hou_Learning_a_Unified_Classifier_Incrementally_via_Rebalancing_CVPR_2019_paper.html |
| http://openaccess.thecvf.com/content_CVPR_2019/papers/Hou_Learning_a_Unified_Classifier_Incrementally_via_Rebalancing_CVPR_2019_paper.pdf | |
| PWC | https://paperswithcode.com/paper/learning-a-unified-classifier-incrementally |
| Repo | https://github.com/hshustc/CVPR19_Incremental_Learning |
| Framework | pytorch |
Partially Encrypted Deep Learning using Functional Encryption
| Title | Partially Encrypted Deep Learning using Functional Encryption |
| Authors | Théo Ryffel, David Pointcheval, Francis Bach, Edouard Dufour-Sans, Romain Gay |
| Abstract | Machine learning on encrypted data has received a lot of attention thanks to recent breakthroughs in homomorphic encryption and secure multi-party computation. It allows outsourcing computation to untrusted servers without sacrificing privacy of sensitive data. We propose a practical framework to perform partially encrypted and privacy-preserving predictions which combines adversarial training and functional encryption. We first present a new functional encryption scheme to efficiently compute quadratic functions so that the data owner controls what can be computed but is not involved in the calculation: it provides a decryption key which allows one to learn a specific function evaluation of some encrypted data. We then show how to use it in machine learning to partially encrypt neural networks with quadratic activation functions at evaluation time and we provide a thorough analysis of the information leaks based on indistinguishability of data items of the same label. Last, since several encryption schemes cannot deal with the last thresholding operation used for classification, we propose a training method to prevent selected sensitive features from leaking which adversarially optimizes the network against an adversary trying to identify these features. This is of great interest for several existing works using partially encrypted machine learning as it comes with almost no cost on the model’s accuracy and significantly improves data privacy. |
| Tasks | |
| Published | 2019-12-01 |
| URL | http://papers.nips.cc/paper/8701-partially-encrypted-deep-learning-using-functional-encryption |
| http://papers.nips.cc/paper/8701-partially-encrypted-deep-learning-using-functional-encryption.pdf | |
| PWC | https://paperswithcode.com/paper/partially-encrypted-deep-learning-using |
| Repo | https://github.com/LaRiffle/collateral-learning |
| Framework | pytorch |
Approximate Bayesian Inference for a Mechanistic Model of Vesicle Release at a Ribbon Synapse
| Title | Approximate Bayesian Inference for a Mechanistic Model of Vesicle Release at a Ribbon Synapse |
| Authors | Cornelius Schröder, Ben James, Leon Lagnado, Philipp Berens |
| Abstract | The inherent noise of neural systems makes it difficult to construct models which accurately capture experimental measurements of their activity. While much research has been done on how to efficiently model neural activity with descriptive models such as linear-nonlinear-models (LN), Bayesian inference for mechanistic models has received considerably less attention. One reason for this is that these models typically lead to intractable likelihoods and thus make parameter inference difficult. Here, we develop an approximate Bayesian inference scheme for a fully stochastic, biophysically inspired model of glutamate release at the ribbon synapse, a highly specialized synapse found in different sensory systems. The model translates known structural features of the ribbon synapse into a set of stochastically coupled equations. We approximate the posterior distributions by updating a parametric prior distribution via Bayesian updating rules and show that model parameters can be efficiently estimated for synthetic and experimental data from in vivo two-photon experiments in the zebrafish retina. Also, we find that the model captures complex properties of the synaptic release such as the temporal precision and outperforms a standard GLM. Our framework provides a viable path forward for linking mechanistic models of neural activity to measured data. |
| Tasks | Bayesian Inference |
| Published | 2019-12-01 |
| URL | http://papers.nips.cc/paper/8929-approximate-bayesian-inference-for-a-mechanistic-model-of-vesicle-release-at-a-ribbon-synapse |
| http://papers.nips.cc/paper/8929-approximate-bayesian-inference-for-a-mechanistic-model-of-vesicle-release-at-a-ribbon-synapse.pdf | |
| PWC | https://paperswithcode.com/paper/approximate-bayesian-inference-for-a |
| Repo | https://github.com/berenslab/abc-ribbon |
| Framework | none |
The MuCoW Test Suite at WMT 2019: Automatically Harvested Multilingual Contrastive Word Sense Disambiguation Test Sets for Machine Translation
| Title | The MuCoW Test Suite at WMT 2019: Automatically Harvested Multilingual Contrastive Word Sense Disambiguation Test Sets for Machine Translation |
| Authors | Aless Raganato, ro, Yves Scherrer, J{"o}rg Tiedemann |
| Abstract | Supervised Neural Machine Translation (NMT) systems currently achieve impressive translation quality for many language pairs. One of the key features of a correct translation is the ability to perform word sense disambiguation (WSD), i.e., to translate an ambiguous word with its correct sense. Existing evaluation benchmarks on WSD capabilities of translation systems rely heavily on manual work and cover only few language pairs and few word types. We present MuCoW, a multilingual contrastive test suite that covers 16 language pairs with more than 200 thousand contrastive sentence pairs, automatically built from word-aligned parallel corpora and the wide-coverage multilingual sense inventory of BabelNet. We evaluate the quality of the ambiguity lexicons and of the resulting test suite on all submissions from 9 language pairs presented in the WMT19 news shared translation task, plus on other 5 language pairs using NMT pretrained models. The MuCoW test suite is available at http://github.com/Helsinki-NLP/MuCoW. |
| Tasks | Machine Translation, Word Sense Disambiguation |
| Published | 2019-08-01 |
| URL | https://www.aclweb.org/anthology/W19-5354/ |
| https://www.aclweb.org/anthology/W19-5354 | |
| PWC | https://paperswithcode.com/paper/the-mucow-test-suite-at-wmt-2019 |
| Repo | https://github.com/Helsinki-NLP/MuCoW |
| Framework | none |
Fast Parallel Algorithms for Statistical Subset Selection Problems
| Title | Fast Parallel Algorithms for Statistical Subset Selection Problems |
| Authors | Sharon Qian, Yaron Singer |
| Abstract | In this paper, we propose a new framework for designing fast parallel algorithms for fundamental statistical subset selection tasks that include feature selection and experimental design. Such tasks are known to be weakly submodular and are amenable to optimization via the standard greedy algorithm. Despite its desirable approximation guarantees, however, the greedy algorithm is inherently sequential and in the worst case, its parallel runtime is linear in the size of the data. Recently, there has been a surge of interest in a parallel optimization technique called adaptive sampling which produces solutions with desirable approximation guarantees for submodular maximization in exponentially faster parallel runtime. Unfortunately, we show that for general weakly submodular functions such accelerations are impossible. The major contribution in this paper is a novel relaxation of submodularity which we call differential submodularity. We first prove that differential submodularity characterizes objectives like feature selection and experimental design. We then design an adaptive sampling algorithm for differentially submodular functions whose parallel runtime is logarithmic in the size of the data and achieves strong approximation guarantees. Through experiments, we show the algorithm’s performance is competitive with state-of-the-art methods and obtains dramatic speedups for feature selection and experimental design problems. |
| Tasks | Feature Selection |
| Published | 2019-12-01 |
| URL | http://papers.nips.cc/paper/8751-fast-parallel-algorithms-for-statistical-subset-selection-problems |
| http://papers.nips.cc/paper/8751-fast-parallel-algorithms-for-statistical-subset-selection-problems.pdf | |
| PWC | https://paperswithcode.com/paper/fast-parallel-algorithms-for-statistical |
| Repo | https://github.com/robo-sq/dash |
| Framework | none |
Neural User Factor Adaptation for Text Classification: Learning to Generalize Across Author Demographics
| Title | Neural User Factor Adaptation for Text Classification: Learning to Generalize Across Author Demographics |
| Authors | Xiaolei Huang, Michael J. Paul |
| Abstract | Language use varies across different demographic factors, such as gender, age, and geographic location. However, most existing document classification methods ignore demographic variability. In this study, we examine empirically how text data can vary across four demographic factors: gender, age, country, and region. We propose a multitask neural model to account for demographic variations via adversarial training. In experiments on four English-language social media datasets, we find that classification performance improves when adapting for user factors. |
| Tasks | Document Classification, Text Classification |
| Published | 2019-06-01 |
| URL | https://www.aclweb.org/anthology/S19-1015/ |
| https://www.aclweb.org/anthology/S19-1015 | |
| PWC | https://paperswithcode.com/paper/neural-user-factor-adaptation-for-text |
| Repo | https://github.com/xiaoleihuang/NUFA |
| Framework | none |
FSA-Net: Learning Fine-Grained Structure Aggregation for Head Pose Estimation From a Single Image
| Title | FSA-Net: Learning Fine-Grained Structure Aggregation for Head Pose Estimation From a Single Image |
| Authors | Tsun-Yi Yang, Yi-Ting Chen, Yen-Yu Lin, Yung-Yu Chuang |
| Abstract | This paper proposes a method for head pose estimation from a single image. Previous methods often predict head poses through landmark or depth estimation and would require more computation than necessary. Our method is based on regression and feature aggregation. For having a compact model, we employ the soft stagewise regression scheme. Existing feature aggregation methods treat inputs as a bag of features and thus ignore their spatial relationship in a feature map. We propose to learn a fine-grained structure mapping for spatially grouping features before aggregation. The fine-grained structure provides part-based information and pooled values. By utilizing learnable and non-learnable importance over the spatial location, different model variants can be generated and form a complementary ensemble. Experiments show that our method outperforms the state-of-the-art methods including both the landmark-free ones and the ones based on landmark or depth estimation. With only a single RGB frame as input, our method even outperforms methods utilizing multi-modality information (RGB-D, RGB-Time) on estimating the yaw angle. Furthermore, the memory overhead of our model is 100 times smaller than those of previous methods. |
| Tasks | Depth Estimation, Head Pose Estimation, Pose Estimation |
| Published | 2019-06-01 |
| URL | http://openaccess.thecvf.com/content_CVPR_2019/html/Yang_FSA-Net_Learning_Fine-Grained_Structure_Aggregation_for_Head_Pose_Estimation_From_CVPR_2019_paper.html |
| http://openaccess.thecvf.com/content_CVPR_2019/papers/Yang_FSA-Net_Learning_Fine-Grained_Structure_Aggregation_for_Head_Pose_Estimation_From_CVPR_2019_paper.pdf | |
| PWC | https://paperswithcode.com/paper/fsa-net-learning-fine-grained-structure |
| Repo | https://github.com/shamangary/FSA-Net |
| Framework | tf |
Improving the Robustness of Question Answering Systems to Question Paraphrasing
| Title | Improving the Robustness of Question Answering Systems to Question Paraphrasing |
| Authors | Wee Chung Gan, Hwee Tou Ng |
| Abstract | Despite the advancement of question answering (QA) systems and rapid improvements on held-out test sets, their generalizability is a topic of concern. We explore the robustness of QA models to question paraphrasing by creating two test sets consisting of paraphrased SQuAD questions. Paraphrased questions from the first test set are very similar to the original questions designed to test QA models{'} over-sensitivity, while questions from the second test set are paraphrased using context words near an incorrect answer candidate in an attempt to confuse QA models. We show that both paraphrased test sets lead to significant decrease in performance on multiple state-of-the-art QA models. Using a neural paraphrasing model trained to generate multiple paraphrased questions for a given source question and a set of paraphrase suggestions, we propose a data augmentation approach that requires no human intervention to re-train the models for improved robustness to question paraphrasing. |
| Tasks | Data Augmentation, Question Answering |
| Published | 2019-07-01 |
| URL | https://www.aclweb.org/anthology/P19-1610/ |
| https://www.aclweb.org/anthology/P19-1610 | |
| PWC | https://paperswithcode.com/paper/improving-the-robustness-of-question |
| Repo | https://github.com/nusnlp/paraphrasing-squad |
| Framework | none |
Adaptive Convolutional Neural Networks
| Title | Adaptive Convolutional Neural Networks |
| Authors | Julio Cesar Zamora, Jesus Adan Cruz Vargas, Omesh Tickoo |
| Abstract | The quest for increased visual recognition performance has led to the development of highly complex neural networks with very deep topologies. To avoid high computing resource requirements of such complex networks and to enable operation on devices with limited resources, this paper introduces adaptive kernels for convolutional layers. Motivated by the non-linear perception response in human visual cells, the input image is used to define the weights of a dynamic kernel called Adaptive kernel. This new adaptive kernel is used to perform a second convolution of the input image generating the output pixel. Adaptive kernels enable accurate recognition with lower memory requirements; This is accomplished through reducing the number of kernels and the number of layers needed in the typical CNN configuration, in addition to reducing the memory used, increasing 2X the training speed and the number of activation function evaluations. Our experiments show a reduction of 70X in the memory used for MNIST, maintaining 99% accuracy and 16X memory reduction for CIFAR10 with 92.5% accuracy. |
| Tasks | |
| Published | 2019-05-01 |
| URL | https://openreview.net/forum?id=ByeWdiR5Ym |
| https://openreview.net/pdf?id=ByeWdiR5Ym | |
| PWC | https://paperswithcode.com/paper/adaptive-convolutional-neural-networks |
| Repo | https://github.com/adapconv/adaptive-cnn |
| Framework | pytorch |
Guided Similarity Separation for Image Retrieval
| Title | Guided Similarity Separation for Image Retrieval |
| Authors | Chundi Liu, Guangwei Yu, Maksims Volkovs, Cheng Chang, Himanshu Rai, Junwei Ma, Satya Krishna Gorti |
| Abstract | Despite recent progress in computer vision, image retrieval remains a challenging open problem. Numerous variations such as view angle, lighting and occlusion make it difficult to design models that are both robust and efficient. Many leading methods traverse the nearest neighbor graph to exploit higher order neighbor information and uncover the highly complex underlying manifold. In this work we propose a different approach where we leverage graph convolutional networks to directly encode neighbor information into image descriptors. We further leverage ideas from clustering and manifold learning, and introduce an unsupervised loss based on pairwise separation of image similarities. Empirically, we demonstrate that our model is able to successfully learn a new descriptor space that significantly improves retrieval accuracy, while still allowing efficient inner product inference. Experiments on five public benchmarks show highly competitive performance with up to 24% relative improvement in mAP over leading baselines. Full code for this work is available here: https://github.com/layer6ai-labs/GSS. |
| Tasks | Image Retrieval |
| Published | 2019-12-01 |
| URL | http://papers.nips.cc/paper/8434-guided-similarity-separation-for-image-retrieval |
| http://papers.nips.cc/paper/8434-guided-similarity-separation-for-image-retrieval.pdf | |
| PWC | https://paperswithcode.com/paper/guided-similarity-separation-for-image |
| Repo | https://github.com/layer6ai-labs/GSS |
| Framework | tf |
Adversarial Training for Weakly Supervised Event Detection
| Title | Adversarial Training for Weakly Supervised Event Detection |
| Authors | Xiaozhi Wang, Xu Han, Zhiyuan Liu, Maosong Sun, Peng Li |
| Abstract | Modern weakly supervised methods for event detection (ED) avoid time-consuming human annotation and achieve promising results by learning from auto-labeled data. However, these methods typically rely on sophisticated pre-defined rules as well as existing instances in knowledge bases for automatic annotation and thus suffer from low coverage, topic bias, and data noise. To address these issues, we build a large event-related candidate set with good coverage and then apply an adversarial training mechanism to iteratively identify those informative instances from the candidate set and filter out those noisy ones. The experiments on two real-world datasets show that our candidate selection and adversarial training can cooperate together to obtain more diverse and accurate training data for ED, and significantly outperform the state-of-the-art methods in various weakly supervised scenarios. The datasets and source code can be obtained from https://github.com/thunlp/Adv-ED. |
| Tasks | |
| Published | 2019-06-01 |
| URL | https://www.aclweb.org/anthology/N19-1105/ |
| https://www.aclweb.org/anthology/N19-1105 | |
| PWC | https://paperswithcode.com/paper/adversarial-training-for-weakly-supervised |
| Repo | https://github.com/thunlp/Adv-ED |
| Framework | pytorch |
Natural Questions: a Benchmark for Question Answering Research
| Title | Natural Questions: a Benchmark for Question Answering Research |
| Authors | Tom Kwiatkowski, Jennimaria Palomaki, Olivia Redfield, Michael Collins, Ankur Parikh, Chris Alberti, Danielle Epstein, Illia Polosukhin, Jacob Devlin, Kenton Lee, Kristina Toutanova, Llion Jones, Matthew Kelcey, Ming-Wei Chang, Andrew M. Dai, Jakob Uszkoreit, Quoc Le, Slav Petrov |
| Abstract | We present the Natural Questions corpus, a question answering dataset. Questions consist of real anonymized, aggregated queries issued to the Google search engine. An annotator is presented with a question along with a Wikipedia page from the top 5 search results, and annotates a long answer (typically a paragraph) and a short answer (one or more entities) if present on the page, or marks null if no long/short answer is present. The public release consists of 307,373 training examples with single annotations, 7,830 examples with 5-way annotations for development data, and a further 7,842 examples 5-way annotated sequestered as test data. We present experiments validating quality of the data. We also describe analysis of 25-way annotations on 302 examples, giving insights into human variability on the annotation task. We introduce robust metrics for the purposes of evaluating question answering systems; demonstrate high human upper bounds on these metrics; and establish baseline results using competitive methods drawn from related literature. |
| Tasks | Question Answering |
| Published | 2019-06-01 |
| URL | https://ai.google/research/pubs/pub47761 |
| https://storage.googleapis.com/pub-tools-public-publication-data/pdf/b8c26e4347adc3453c15d96a09e6f7f102293f71.pdf | |
| PWC | https://paperswithcode.com/paper/natural-questions-a-benchmark-for-question |
| Repo | https://github.com/google-research/language |
| Framework | tf |
Eidetic 3D LSTM: A Model for Video Prediction and Beyond
| Title | Eidetic 3D LSTM: A Model for Video Prediction and Beyond |
| Authors | Yunbo Wang, Lu Jiang, Ming-Hsuan Yang, Li-Jia Li, Mingsheng Long, Li Fei-Fei |
| Abstract | Spatiotemporal predictive learning, though long considered to be a promising self-supervised feature learning method, seldom shows its effectiveness beyond future video prediction. The reason is that it is difficult to learn good representations for both short-term frame dependency and long-term high-level relations. We present a new model, Eidetic 3D LSTM (E3D-LSTM), that integrates 3D convolutions into RNNs. The encapsulated 3D-Conv makes local perceptrons of RNNs motion-aware and enables the memory cell to store better short-term features. For long-term relations, we make the present memory state interact with its historical records via a gate-controlled self-attention module. We describe this memory transition mechanism eidetic as it is able to effectively recall the stored memories across multiple time stamps even after long periods of disturbance. We first evaluate the E3D-LSTM network on widely-used future video prediction datasets and achieve the state-of-the-art performance. Then we show that the E3D-LSTM network also performs well on the early activity recognition to infer what is happening or what will happen after observing only limited frames of video. This task aligns well with video prediction, as action intentions and tendency are important to achieve the state-of-the-art performance. |
| Tasks | Activity Recognition, Video Prediction |
| Published | 2019-05-01 |
| URL | https://openreview.net/forum?id=B1lKS2AqtX |
| https://openreview.net/pdf?id=B1lKS2AqtX | |
| PWC | https://paperswithcode.com/paper/eidetic-3d-lstm-a-model-for-video-prediction |
| Repo | https://github.com/google/e3d_lstm |
| Framework | tf |
Paradoxes in Fair Machine Learning
| Title | Paradoxes in Fair Machine Learning |
| Authors | Paul Goelz, Anson Kahng, Ariel D. Procaccia |
| Abstract | Equalized odds is a statistical notion of fairness in machine learning that ensures that classification algorithms do not discriminate against protected groups. We extend equalized odds to the setting of cardinality-constrained fair classification, where we have a bounded amount of a resource to distribute. This setting coincides with classic fair division problems, which allows us to apply concepts from that literature in parallel to equalized odds. In particular, we consider the axioms of resource monotonicity, consistency, and population monotonicity, all three of which relate different allocation instances to prevent paradoxes. Using a geometric characterization of equalized odds, we examine the compatibility of equalized odds with these axioms. We empirically evaluate the cost of allocation rules that satisfy both equalized odds and axioms of fair division on a dataset of FICO credit scores. |
| Tasks | |
| Published | 2019-12-01 |
| URL | http://papers.nips.cc/paper/9043-paradoxes-in-fair-machine-learning |
| http://papers.nips.cc/paper/9043-paradoxes-in-fair-machine-learning.pdf | |
| PWC | https://paperswithcode.com/paper/paradoxes-in-fair-machine-learning |
| Repo | https://github.com/pgoelz/equalized |
| Framework | none |
AndroShield: Automated Android Applications Vulnerability Detection, a Hybrid Static and Dynamic Analysis Approach
| Title | AndroShield: Automated Android Applications Vulnerability Detection, a Hybrid Static and Dynamic Analysis Approach |
| Authors | Amr Amin, Amgad Eldessouki, Menna Tullah Magdy, Nouran Abdeen, Hanan Hindy, Islam Hegazy |
| Abstract | The security of mobile applications has become a major research field which is associated with a lot of challenges. The high rate of developing mobile applications has resulted in less secure applications. This is due to what is called the “rush to release” as defined by Ponemon Institute. Security testing—which is considered one of the main phases of the development life cycle—is either not performed or given minimal time; hence, there is a need for security testing automation. One of the techniques used is Automated Vulnerability Detection. Vulnerability detection is one of the security tests that aims at pinpointing potential security leaks. Fixing those leaks results in protecting smart-phones and tablet mobile device users against attacks. This paper focuses on building a hybrid approach of static and dynamic analysis for detecting the vulnerabilities of Android applications. This approach is capsuled in a usable platform (web application) to make it easy to use for both public users and professional developers. Static analysis, on one hand, performs code analysis. It does not require running the application to detect vulnerabilities. Dynamic analysis, on the other hand, detects the vulnerabilities that are dependent on the run-time behaviour of the application and cannot be detected using static analysis. The model is evaluated against different applications with different security vulnerabilities. Compared with other detection platforms, our model detects information leaks as well as insecure network requests alongside other commonly detected flaws that harm users’ privacy. The code is available through a GitHub repository for public contribution. |
| Tasks | Anomaly Detection, Mobile Security, Vulnerability Detection |
| Published | 2019-10-22 |
| URL | https://www.mdpi.com/2078-2489/10/10/326 |
| https://www.mdpi.com/2078-2489/10/10/326/pdf | |
| PWC | https://paperswithcode.com/paper/androshield-automated-android-applications |
| Repo | https://github.com/AmrAshraf/AndroShield |
| Framework | none |
