Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

"Topic": models, code, and papers

Cats or CAT scans: transfer learning from natural or medical image source datasets?

Oct 12, 2018
Veronika Cheplygina

Transfer learning is a widely used strategy in medical image analysis. Instead of only training a network with a limited amount of data from the target task of interest, we can first train the network with other, potentially larger source datasets, creating a more robust model. The source datasets do not have to be related to the target task. For a classification task in lung CT images, we could use both head CT images, or images of cats, as the source. While head CT images appear more similar to lung CT images, the number and diversity of cat images might lead to a better model overall. In this survey we review a number of papers that have performed similar comparisons. Although the answer to which strategy is best seems to be "it depends", we discuss a number of research directions we need to take as a community, to gain more understanding of this topic.

  Access Paper or Ask Questions

Learning Goal-Oriented Visual Dialog via Tempered Policy Gradient

Oct 04, 2018
Rui Zhao, Volker Tresp

Learning goal-oriented dialogues by means of deep reinforcement learning has recently become a popular research topic. However, commonly used policy-based dialogue agents often end up focusing on simple utterances and suboptimal policies. To mitigate this problem, we propose a class of novel temperature-based extensions for policy gradient methods, which are referred to as Tempered Policy Gradients (TPGs). On a recent AI-testbed, i.e., the GuessWhat?! game, we achieve significant improvements with two innovations. The first one is an extension of the state-of-the-art solutions with Seq2Seq and Memory Network structures that leads to an improvement of 7%. The second one is the application of our newly developed TPG methods, which improves the performance additionally by around 5% and, even more importantly, helps produce more convincing utterances.

* Accepted by IEEE Spoken Language Technology (SLT 2018), Athens, Greece 

  Access Paper or Ask Questions

Neural Argument Generation Augmented with Externally Retrieved Evidence

May 25, 2018
Xinyu Hua, Lu Wang

High quality arguments are essential elements for human reasoning and decision-making processes. However, effective argument construction is a challenging task for both human and machines. In this work, we study a novel task on automatically generating arguments of a different stance for a given statement. We propose an encoder-decoder style neural network-based argument generation model enriched with externally retrieved evidence from Wikipedia. Our model first generates a set of talking point phrases as intermediate representation, followed by a separate decoder producing the final argument based on both input and the keyphrases. Experiments on a large-scale dataset collected from Reddit show that our model constructs arguments with more topic-relevant content than a popular sequence-to-sequence generation model according to both automatic evaluation and human assessments.

* This paper is accepted as to ACL 2018 

  Access Paper or Ask Questions

Unsupervised object discovery for instance recognition

Jan 24, 2018
Oriane Siméoni, Ahmet Iscen, Giorgos Tolias, Yannis Avrithis, Ondrej Chum

Severe background clutter is challenging in many computer vision tasks, including large-scale image retrieval. Global descriptors, that are popular due to their memory and search efficiency, are especially prone to corruption by such a clutter. Eliminating the impact of the clutter on the image descriptor increases the chance of retrieving relevant images and prevents topic drift due to actually retrieving the clutter in the case of query expansion. In this work, we propose a novel salient region detection method. It captures, in an unsupervised manner, patterns that are both discriminative and common in the dataset. Saliency is based on a centrality measure of a nearest neighbor graph constructed from regional CNN representations of dataset images. The descriptors derived from the salient regions improve particular object retrieval, most noticeably in a large collections containing small objects.

  Access Paper or Ask Questions

A Deep Reinforcement Learning Chatbot (Short Version)

Jan 20, 2018
Iulian V. Serban, Chinnadhurai Sankar, Mathieu Germain, Saizheng Zhang, Zhouhan Lin, Sandeep Subramanian, Taesup Kim, Michael Pieper, Sarath Chandar, Nan Rosemary Ke, Sai Rajeswar, Alexandre de Brebisson, Jose M. R. Sotelo, Dendi Suhubdy, Vincent Michalski, Alexandre Nguyen, Joelle Pineau, Yoshua Bengio

We present MILABOT: a deep reinforcement learning chatbot developed by the Montreal Institute for Learning Algorithms (MILA) for the Amazon Alexa Prize competition. MILABOT is capable of conversing with humans on popular small talk topics through both speech and text. The system consists of an ensemble of natural language generation and retrieval models, including neural network and template-based models. By applying reinforcement learning to crowdsourced data and real-world user interactions, the system has been trained to select an appropriate response from the models in its ensemble. The system has been evaluated through A/B testing with real-world users, where it performed significantly better than other systems. The results highlight the potential of coupling ensemble systems with deep reinforcement learning as a fruitful path for developing real-world, open-domain conversational agents.

* 9 pages, 1 figure, 2 tables; presented at NIPS 2017, Conversational AI: "Today's Practice and Tomorrow's Potential" Workshop 

  Access Paper or Ask Questions

Aerial Drop of Robots and Sensors for Optimal Area Coverage

Nov 28, 2017
Kostas Alexis

The problem of rapid optimal coverage through the distribution a team of robots or static sensors via means of aerial drop is the topic of this work. Considering a nonholonomic (fixed-wing) aerial robot that corresponds to the carrier of a set of small holonomic (rotorcraft) aerial robots as well as static modules that are all equipped with a camera sensor, we address the problem of selecting optimal aerial drop times and configurations while the motion capabilities of the small aerial robots are also exploited to further survey their area of responsibility until they hit the ground. The overall solution framework consists of lightweight path-planning algorithms that can run on virtually any processing unit that might be available on-board. Evaluation studies in simulation as well as a set of elementary experiments that prove the validity of important assumptions illustrate the potential of the approach.

* 6 pages, 9 figures, technical report 

  Access Paper or Ask Questions

Jointly Trained Sequential Labeling and Classification by Sparse Attention Neural Networks

Sep 28, 2017
Mingbo Ma, Kai Zhao, Liang Huang, Bing Xiang, Bowen Zhou

Sentence-level classification and sequential labeling are two fundamental tasks in language understanding. While these two tasks are usually modeled separately, in reality, they are often correlated, for example in intent classification and slot filling, or in topic classification and named-entity recognition. In order to utilize the potential benefits from their correlations, we propose a jointly trained model for learning the two tasks simultaneously via Long Short-Term Memory (LSTM) networks. This model predicts the sentence-level category and the word-level label sequence from the stepwise output hidden representations of LSTM. We also introduce a novel mechanism of "sparse attention" to weigh words differently based on their semantic relevance to sentence-level classification. The proposed method outperforms baseline models on ATIS and TREC datasets.

* interspeech 2017 

  Access Paper or Ask Questions

Understanding Infographics through Textual and Visual Tag Prediction

Sep 26, 2017
Zoya Bylinskii, Sami Alsheikh, Spandan Madan, Adria Recasens, Kimberli Zhong, Hanspeter Pfister, Fredo Durand, Aude Oliva

We introduce the problem of visual hashtag discovery for infographics: extracting visual elements from an infographic that are diagnostic of its topic. Given an infographic as input, our computational approach automatically outputs textual and visual elements predicted to be representative of the infographic content. Concretely, from a curated dataset of 29K large infographic images sampled across 26 categories and 391 tags, we present an automated two step approach. First, we extract the text from an infographic and use it to predict text tags indicative of the infographic content. And second, we use these predicted text tags as a supervisory signal to localize the most diagnostic visual elements from within the infographic i.e. visual hashtags. We report performances on a categorization and multi-label tag prediction problem and compare our proposed visual hashtags to human annotations.

  Access Paper or Ask Questions

The Convergence of Machine Learning and Communications

Aug 28, 2017
Wojciech Samek, Slawomir Stanczak, Thomas Wiegand

The areas of machine learning and communication technology are converging. Today's communications systems generate a huge amount of traffic data, which can help to significantly enhance the design and management of networks and communication components when combined with advanced machine learning methods. Furthermore, recently developed end-to-end training procedures offer new ways to jointly optimize the components of a communication system. Also in many emerging application fields of communication technology, e.g., smart cities or internet of things, machine learning methods are of central importance. This paper gives an overview over the use of machine learning in different areas of communications and discusses two exemplar applications in wireless networking. Furthermore, it identifies promising future research topics and discusses their potential impact.

* 8 pages, 4 figures 

  Access Paper or Ask Questions

Exploring text datasets by visualizing relevant words

Jul 17, 2017
Franziska Horn, Leila Arras, Grégoire Montavon, Klaus-Robert Müller, Wojciech Samek

When working with a new dataset, it is important to first explore and familiarize oneself with it, before applying any advanced machine learning algorithms. However, to the best of our knowledge, no tools exist that quickly and reliably give insight into the contents of a selection of documents with respect to what distinguishes them from other documents belonging to different categories. In this paper we propose to extract `relevant words' from a collection of texts, which summarize the contents of documents belonging to a certain class (or discovered cluster in the case of unlabeled datasets), and visualize them in word clouds to allow for a survey of salient features at a glance. We compare three methods for extracting relevant words and demonstrate the usefulness of the resulting word clouds by providing an overview of the classes contained in a dataset of scientific publications as well as by discovering trending topics from recent New York Times article snippets.

  Access Paper or Ask Questions