Paper Group NAWR 21
Multi-Task Learning for Document Ranking and Query Suggestion. Image Deblurring with a Class-Specific Prior. Identifying Key Sentences for Precision Oncology Using Semi-Supervised Learning. Introducing a Lexicon of Verbal Polarity Shifters for English. Fast Adaptive RNN Encoder-Decoder for Anomaly Detection in SMD Assembly Machine. Deep-learning-as …
Multi-Task Learning for Document Ranking and Query Suggestion
| Title | Multi-Task Learning for Document Ranking and Query Suggestion |
| Authors | Wasi Uddin Ahmad, Kai-Wei Chang, Hongning Wang |
| Abstract | We propose a multi-task learning framework to jointly learn document ranking and query suggestion for web search. It consists of two major components, a document ranker, and a query recommender. Document ranker combines current query and session information and compares the combined representation with document representation to rank the documents. Query recommender tracks users’ query reformulation sequence considering all previous in-session queries using a sequence to sequence approach. As both tasks are driven by the users’ underlying search intent, we perform joint learning of these two components through session recurrence, which encodes search context and intent. Extensive comparisons against state-of-the-art document ranking and query suggestion algorithms are performed on the public AOL search log, and the promising results endorse the effectiveness of the joint learning framework. |
| Tasks | Document Ranking, Multi-Task Learning |
| Published | 2018-01-01 |
| URL | https://openreview.net/forum?id=SJ1nzBeA- |
| https://openreview.net/pdf?id=SJ1nzBeA- | |
| PWC | https://paperswithcode.com/paper/multi-task-learning-for-document-ranking-and |
| Repo | https://github.com/wasiahmad/mnsrf_ranking_suggestion |
| Framework | pytorch |
Image Deblurring with a Class-Specific Prior
| Title | Image Deblurring with a Class-Specific Prior |
| Authors | Saeed Anwar ; Cong Phuoc Huynh ; Fatih Porikli |
| Abstract | A fundamental problem in image deblurring is to recover reliably distinct spatial frequencies that have been suppressed by the blur kernel. To tackle this issue, existing image deblurring techniques often rely on generic image priors such as the sparsity of salient features including image gradients and edges. However, these priors only help recover part of the frequency spectrum, such as the frequencies near the high-end. To this end, we pose the following specific questions: (i) Does any image class information offer an advantage over existing generic priors for image quality restoration? (ii) If a class-specific prior exists, how should it be encoded into a deblurring framework to recover attenuated image frequencies? Throughout this work, we devise a class-specific prior based on the band-pass filter responses and incorporate it into a deblurring strategy. More specifically, we show that the subspace of band-pass filtered images and their intensity distributions serve as useful priors for recovering image frequencies that are difficult to recover by generic image priors. We demonstrate that our image deblurring framework, when equipped with the above priors, significantly outperforms many state-of-the-art methods using generic image priors or class-specific exemplars. |
| Tasks | Deblurring |
| Published | 2018-07-11 |
| URL | https://ieeexplore.ieee.org/document/8409975 |
| https://ieeexplore.ieee.org/document/8409975 | |
| PWC | https://paperswithcode.com/paper/image-deblurring-with-a-class-specific-prior |
| Repo | https://github.com/saeed-anwar/Class_Specific_Deblurring |
| Framework | none |
Identifying Key Sentences for Precision Oncology Using Semi-Supervised Learning
| Title | Identifying Key Sentences for Precision Oncology Using Semi-Supervised Learning |
| Authors | Jurica {\v{S}}eva, Martin Wackerbauer, Ulf Leser |
| Abstract | We present a machine learning pipeline that identifies key sentences in abstracts of oncological articles to aid evidence-based medicine. This problem is characterized by the lack of gold standard datasets, data imbalance and thematic differences between available silver standard corpora. Additionally, available training and target data differs with regard to their domain (professional summaries vs. sentences in abstracts). This makes supervised machine learning inapplicable. We propose the use of two semi-supervised machine learning approaches: To mitigate difficulties arising from heterogeneous data sources, overcome data imbalance and create reliable training data we propose using transductive learning from positive and unlabelled data (PU Learning). For obtaining a realistic classification model, we propose the use of abstracts summarised in relevant sentences as unlabelled examples through Self-Training. The best model achieves 84{%} accuracy and 0.84 F1 score on our dataset |
| Tasks | |
| Published | 2018-07-01 |
| URL | https://www.aclweb.org/anthology/W18-2305/ |
| https://www.aclweb.org/anthology/W18-2305 | |
| PWC | https://paperswithcode.com/paper/identifying-key-sentences-for-precision |
| Repo | https://github.com/nachne/semisuper |
| Framework | none |
Introducing a Lexicon of Verbal Polarity Shifters for English
| Title | Introducing a Lexicon of Verbal Polarity Shifters for English |
| Authors | Marc Schulder, Michael Wiegand, Josef Ruppenhofer, Stephanie Köser |
| Abstract | |
| Tasks | Natural Language Inference, Relation Extraction, Sentiment Analysis |
| Published | 2018-05-01 |
| URL | https://www.aclweb.org/anthology/papers/L18-1222/l18-1222 |
| https://www.aclweb.org/anthology/L18-1222 | |
| PWC | https://paperswithcode.com/paper/introducing-a-lexicon-of-verbal-polarity |
| Repo | https://github.com/marcschulder/lrec2018 |
| Framework | none |
Fast Adaptive RNN Encoder-Decoder for Anomaly Detection in SMD Assembly Machine
| Title | Fast Adaptive RNN Encoder-Decoder for Anomaly Detection in SMD Assembly Machine |
| Authors | YeongHyeon Park, Il Dong Yun |
| Abstract | Surface Mounted Device (SMD) assembly machine manufactures various products on a flexible manufacturing line. An anomaly detection model that can adapt to the various manufacturing environments very fast is required. In this paper, we proposed a fast adaptive anomaly detection model based on a Recurrent Neural Network (RNN) Encoder–Decoder with operating machine sounds. RNN Encoder–Decoder has a structure very similar to Auto-Encoder (AE), but the former has significantly reduced parameters compared to the latter because of its rolled structure. Thus, the RNN Encoder–Decoder only requires a short training process for fast adaptation. The anomaly detection model decides abnormality based on Euclidean distance between generated sequences and observed sequence from machine sounds. Experimental evaluation was conducted on a set of dataset from the SMD assembly machine. Results showed cutting-edge performance with fast adaptation. View Full-Text |
| Tasks | Anomaly Detection |
| Published | 2018-10-22 |
| URL | https://www.mdpi.com/1424-8220/18/10/3573 |
| https://www.mdpi.com/1424-8220/18/10/3573/pdf | |
| PWC | https://paperswithcode.com/paper/fast-adaptive-rnn-encoder-decoder-for-anomaly |
| Repo | https://github.com/YeongHyeon/FARED_for_Anomaly_Detection |
| Framework | tf |
Deep-learning-assisted diagnosis for knee magnetic resonance imaging: Development and retrospective validation of MRNet
| Title | Deep-learning-assisted diagnosis for knee magnetic resonance imaging: Development and retrospective validation of MRNet |
| Authors | Nicholas Bien, Pranav Rajpurkar, Robyn L. Ball, Jeremy Irvin, Allison Park, Erik Jones, Michael Bereket, Bhavik N. Patel, Kristen W. Yeom, Katie Shpanskaya, Safwan Halabi, Evan Zucker, Gary Fanton, Derek F. Amanatullah, Christopher F. Beaulieu, Geoffrey M. Riley, Russell J. Stewart, Francis G. Blankenberg, David B. Larson, Ricky H. Jones, Curtis P. Langlotz, Andrew Y. Ng, Matthew P. Lungren |
| Abstract | Magnetic resonance imaging (MRI) of the knee is the preferred method for diagnosing knee injuries. However, interpretation of knee MRI is time-intensive and subject to diagnostic error and variability. An automated system for interpreting knee MRI could prioritize highrisk patients and assist clinicians in making diagnoses. Deep learning methods, in being able to automatically learn layers of features, are well suited for modeling the complex relationships between medical images and their interpretations. In this study we developed a deep learning model for detecting general abnormalities and specific diagnoses (anterior cruciate ligament [ACL] tears and meniscal tears) on knee MRI exams. We then measured the effect of providing the model’s predictions to clinical experts during interpretation. |
| Tasks | |
| Published | 2018-11-27 |
| URL | https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1002699 |
| https://journals.plos.org/plosmedicine/article/file?id=10.1371/journal.pmed.1002699&type=printable | |
| PWC | https://paperswithcode.com/paper/deep-learning-assisted-diagnosis-for-knee |
| Repo | https://github.com/ahmedbesbes/mrnet |
| Framework | pytorch |
Accelerated Spectral Ranking
| Title | Accelerated Spectral Ranking |
| Authors | Arpit Agarwal, Prathamesh Patil, Shivani Agarwal |
| Abstract | The problem of rank aggregation from pairwise and multiway comparisons has a wide range of implications, ranging from recommendation systems to sports rankings to social choice. Some of the most popular algorithms for this problem come from the class of spectral ranking algorithms; these include the rank centrality (RC) algorithm for pairwise comparisons, which returns consistent estimates under the Bradley-Terry-Luce (BTL) model for pairwise comparisons (Negahban et al., 2017), and its generalization, the Luce spectral ranking (LSR) algorithm, which returns consistent estimates under the more general multinomial logit (MNL) model for multiway comparisons (Maystre & Grossglauser, 2015). In this paper, we design a provably faster spectral ranking algorithm, which we call accelerated spectral ranking (ASR), that is also consistent under the MNL/BTL models. Our accelerated algorithm is achieved by designing a random walk that has a faster mixing time than the random walks associated with previous algorithms. In addition to a faster algorithm, our results yield improved sample complexity bounds for recovery of the MNL/BTL parameters: to the best of our knowledge, we give the first general sample complexity bounds for recovering the parameters of the MNL model from multiway comparisons under any (connected) comparison graph (and improve significantly over previous bounds for the BTL model for pairwise comparisons). We also give a message-passing interpretation of our algorithm, which suggests a decentralized distributed implementation. Our experiments on several real-world and synthetic datasets confirm that our new ASR algorithm is indeed orders of magnitude faster than existing algorithms. |
| Tasks | Recommendation Systems |
| Published | 2018-07-01 |
| URL | https://icml.cc/Conferences/2018/Schedule?showEvent=1999 |
| http://proceedings.mlr.press/v80/agarwal18b/agarwal18b.pdf | |
| PWC | https://paperswithcode.com/paper/accelerated-spectral-ranking |
| Repo | https://github.com/agarpit/asr |
| Framework | none |
FireSim: FPGA-Accelerated Cycle-Exact Scale-Out System Simulation in the Public Cloud
| Title | FireSim: FPGA-Accelerated Cycle-Exact Scale-Out System Simulation in the Public Cloud |
| Authors | Sagar Karandikar, Howard Mao, Donggyu Kim, David Biancolin, Alon Amid, Dayeol Lee, Nathan Pemberton, Emmanuel Amaro, Colin Schmidt, Aditya Chopra, Qijing Huang, Kyle Kovacs, Borivoje Nikolic, Randy Katz, Jonathan Bachrach, Krste Asanovic |
| Abstract | We present FireSim, an open-source simulation platform that enables cycle-exact microarchitectural simulation of large scale-out clusters by combining FPGA-accelerated simulation of silicon-proven RTL designs with a scalable, distributed network simulation. Unlike prior FPGA-accelerated simulation tools, FireSim runs on Amazon EC2 F1, a public cloud FPGA platform, which greatly improves usability, provides elasticity, and lowers the cost of large-scale FPGAbased experiments. We describe the design and implementation of FireSim and show how it can provide sufficient performance to run modern applications at scale, to enable true hardware-software co-design. As an example, we demonstrate automatically generating and deploying a target cluster of 1,024 3.2 GHz quad-core server nodes, each with 16 GB of DRAM, interconnected by a 200 Gbit/s network with 2 microsecond latency, which simulates at a 3.4 MHz processor clock rate (less than 1,000x slowdown over real-time). In aggregate, this FireSim instantiation simulates 4,096 cores and 16 TB of memory, runs ˜14 billion instructions per second, and harnesses 12.8 million dollars worth of FPGAs—at a total cost of only ˜$100 per simulation hour to the user. We present several examples to show how FireSim can be used to explore various research directions in warehouse-scale machine design, including modeling networks with high-bandwidth and low-latency, integrating arbitrary RTL designs for a variety of commodity and specialized datacenter nodes, and modeling a variety of datacenter organizations, as well as reusing the scale-out FireSim infrastructure to enable fast, massively parallel cycle-exact single-node microarchitectural experimentation. |
| Tasks | |
| Published | 2018-06-02 |
| URL | https://dl.acm.org/citation.cfm?id=3276543 |
| https://sagark.org/assets/pubs/firesim-isca2018.pdf | |
| PWC | https://paperswithcode.com/paper/firesim-fpga-accelerated-cycle-exact-scale |
| Repo | https://github.com/firesim/firesim |
| Framework | none |
Attention in Convolutional LSTM for Gesture Recognition
| Title | Attention in Convolutional LSTM for Gesture Recognition |
| Authors | Liang Zhang, Guangming Zhu, Lin Mei, Peiyi Shen, Syed Afaq Ali Shah, Mohammed Bennamoun |
| Abstract | Convolutional long short-term memory (LSTM) networks have been widely used for action/gesture recognition, and different attention mechanisms have also been embedded into the LSTM or the convolutional LSTM (ConvLSTM) networks. Based on the previous gesture recognition architectures which combine the three-dimensional convolution neural network (3DCNN) and ConvLSTM, this paper explores the effects of attention mechanism in ConvLSTM. Several variants of ConvLSTM are evaluated: (a) Removing the convolutional structures of the three gates in ConvLSTM, (b) Applying the attention mechanism on the input of ConvLSTM, (c) Reconstructing the input and (d) output gates respectively with the modified channel-wise attention mechanism. The evaluation results demonstrate that the spatial convolutions in the three gates scarcely contribute to the spatiotemporal feature fusion, and the attention mechanisms embedded into the input and output gates cannot improve the feature fusion. In other words, ConvLSTM mainly contributes to the temporal fusion along with the recurrent steps to learn the long-term spatiotemporal features, when taking as input the spatial or spatiotemporal features. On this basis, a new variant of LSTM is derived, in which the convolutional structures are only embedded into the input-to-state transition of LSTM. The code of the LSTM variants is publicly available. |
| Tasks | Gesture Recognition |
| Published | 2018-12-01 |
| URL | http://papers.nips.cc/paper/7465-attention-in-convolutional-lstm-for-gesture-recognition |
| http://papers.nips.cc/paper/7465-attention-in-convolutional-lstm-for-gesture-recognition.pdf | |
| PWC | https://paperswithcode.com/paper/attention-in-convolutional-lstm-for-gesture |
| Repo | https://github.com/GuangmingZhu/AttentionConvLSTM |
| Framework | tf |
Studying Muslim Stereotyping through Microportrait Extraction
| Title | Studying Muslim Stereotyping through Microportrait Extraction |
| Authors | Antske Fokkens, Nel Ruigrok, Camiel Beukeboom, Gagestein Sarah, Wouter van Atteveldt |
| Abstract | |
| Tasks | |
| Published | 2018-05-01 |
| URL | https://www.aclweb.org/anthology/L18-1590/ |
| https://www.aclweb.org/anthology/L18-1590 | |
| PWC | https://paperswithcode.com/paper/studying-muslim-stereotyping-through |
| Repo | https://github.com/cltl/micro-portraits |
| Framework | none |
Training and Adapting Multilingual NMT for Less-resourced and Morphologically Rich Languages
| Title | Training and Adapting Multilingual NMT for Less-resourced and Morphologically Rich Languages |
| Authors | Mat{=\i}ss Rikters, M{=a}rcis Pinnis, Rihards Kri{\v{s}}lauks |
| Abstract | |
| Tasks | Machine Translation |
| Published | 2018-05-01 |
| URL | https://www.aclweb.org/anthology/L18-1595/ |
| https://www.aclweb.org/anthology/L18-1595 | |
| PWC | https://paperswithcode.com/paper/training-and-adapting-multilingual-nmt-for |
| Repo | https://github.com/tilde-nlp/multilingual-nmt-data-prep |
| Framework | none |
mlpack 3: a fast, flexible machine learning library
| Title | mlpack 3: a fast, flexible machine learning library |
| Authors | Ryan R. Curtin, Marcus Edel, Mikhail Lozhnikov, Yannis Mentekidis, Sumedh Ghaisas, Shangtong Zhang |
| Abstract | In the past several years, the field of machine learning has seen an explosion of interest and excitement, with hundreds or thousands of algorithms developed for different tasks every year. But a primary problem faced by the field is the ability to scale to larger and larger data – since it is known that training on larger datasets typically produces better results. Therefore, the development of new algorithms for the continued growth of the field depends largely on the existence of good tooling and libraries that enable researchers and practitioners to quickly prototype and develop solutions. Simultaneously, useful libraries must also be efficient and well-implemented. This has motivated our development of mlpack. mlpack is a flexible and fast machine learning library written in C++ that has bindings that allow use from the command-line and from Python, with support for other languages in active development. mlpack has been developed actively for over 10 years, with over 100 contributors from around the world, and is a frequent mentoring organization in the Google Summer of Code program. If used in C++, the library allows flexibility with no speed penalty through policy-based design and template metaprogramming; but bindings are available to other languages, which allow easy use of the fast mlpack codebase. For fast linear algebra, mlpack is built on the Armadillo C++ matrix library, which in turn can use an optimized BLAS implementation such as OpenBLAS or even NVBLAS which would allow mlpack algorithms to be run on the GPU. In order to provide fast code, template metaprogramming is used throughout the library to reduce runtime overhead by performing any possible computations and optimizations at compile time. An automatic benchmarking system is developed and used to test the efficiency of mlpack’s algorithms. mlpack contains a number of standard machine learning algorithms, such as logistic regression, random forests, and k-means clustering, and also contains cutting-edge techniques such as a compile-time optimized deep learning and reinforcement learning framework, dual-tree algorithms for nearest neighbor search and other tasks, a generic optimization framework with numerous optimizers, a generic hyper-parameter tuner, and other recently published machine learning algorithms. For a more comprehensive introduction to mlpack, see the website at http://www.mlpack.org/ |
| Tasks | |
| Published | 2018-06-18 |
| URL | https://doi.org/10.21105/joss.00726 |
| https://doi.org/10.21105/joss.00726 | |
| PWC | https://paperswithcode.com/paper/mlpack-3-a-fast-flexible-machine-learning |
| Repo | https://github.com/mlpack/mlpack |
| Framework | none |
Memory Replay GANs: Learning to Generate New Categories without Forgetting
| Title | Memory Replay GANs: Learning to Generate New Categories without Forgetting |
| Authors | Chenshen Wu, Luis Herranz, Xialei Liu, Yaxing Wang, Joost Van De Weijer, Bogdan Raducanu |
| Abstract | Previous works on sequential learning address the problem of forgetting in discriminative models. In this paper we consider the case of generative models. In particular, we investigate generative adversarial networks (GANs) in the task of learning new categories in a sequential fashion. We first show that sequential fine tuning renders the network unable to properly generate images from previous categories (i.e. forgetting). Addressing this problem, we propose Memory Replay GANs (MeRGANs), a conditional GAN framework that integrates a memory replay generator. We study two methods to prevent forgetting by leveraging these replays, namely joint training with replay and replay alignment. Qualitative and quantitative experimental results in MNIST, SVHN and LSUN datasets show that our memory replay approach can generate competitive images while significantly mitigating the forgetting of previous categories. |
| Tasks | |
| Published | 2018-12-01 |
| URL | http://papers.nips.cc/paper/7836-memory-replay-gans-learning-to-generate-new-categories-without-forgetting |
| http://papers.nips.cc/paper/7836-memory-replay-gans-learning-to-generate-new-categories-without-forgetting.pdf | |
| PWC | https://paperswithcode.com/paper/memory-replay-gans-learning-to-generate-new |
| Repo | https://github.com/WuChenshen/MeRGAN |
| Framework | tf |
A review of Spanish corpora annotated with negation
| Title | A review of Spanish corpora annotated with negation |
| Authors | Salud Mar{'\i}a Jim{'e}nez-Zafra, Roser Morante, Maite Martin, L. Alfonso Ure{~n}a-L{'o}pez |
| Abstract | The availability of corpora annotated with negation information is essential to develop negation processing systems in any language. However, there is a lack of these corpora even for languages like English, and when there are corpora available they are small and the annotations are not always compatible across corpora. In this paper we review the existing corpora annotated with negation in Spanish with the purpose of first, gathering the information to make it available for other researchers and, second, analyzing how compatible are the corpora and how has the linguistic phenomenon been addressed. Our final aim is to develop a supervised negation processing system for Spanish, for which we need training and test data. Our analysis shows that it will not be possible to merge the small corpora existing for Spanish due to lack of compatibility in the annotations. |
| Tasks | Question Answering, Sentiment Analysis |
| Published | 2018-08-01 |
| URL | https://www.aclweb.org/anthology/C18-1078/ |
| https://www.aclweb.org/anthology/C18-1078 | |
| PWC | https://paperswithcode.com/paper/a-review-of-spanish-corpora-annotated-with |
| Repo | https://github.com/sjzafra/spanish_negation_corpora |
| Framework | none |
Data Anonymization for Requirements Quality Analysis: a Reproducible Automatic Error Detection Task
| Title | Data Anonymization for Requirements Quality Analysis: a Reproducible Automatic Error Detection Task |
| Authors | Juyeon Kang, Jungyeul Park |
| Abstract | |
| Tasks | |
| Published | 2018-05-01 |
| URL | https://www.aclweb.org/anthology/L18-1702/ |
| https://www.aclweb.org/anthology/L18-1702 | |
| PWC | https://paperswithcode.com/paper/data-anonymization-for-requirements-quality |
| Repo | https://github.com/jungyeul/rqa |
| Framework | none |
