• Publications
  • Influence
EPOpt: Learning Robust Neural Network Policies Using Model Ensembles
The EPOpt algorithm is introduced, which uses an ensemble of simulated source domains and a form of adversarial training to learn policies that are robust and generalize to a broad range of possible target domains, including unmodeled effects. Expand
An Autoencoder Approach to Learning Bilingual Word Representations
This work explores the use of autoencoder-based methods for cross-language learning of vectorial word representations that are coherent between two languages, while not relying on word-level alignments, and achieves state-of-the-art performance. Expand
Diversity driven attention model for query-based abstractive summarization
This work proposes a model for the query-based summarization task based on the encode-attend-decode paradigm with two key additions: a query attention model which learns to focus on different portions of the query at different time steps and a new diversity based Attention model which aims to alleviate the problem of repeating phrases in the summary. Expand
Correlational Neural Networks
This work proposes an AE-based approach, correlational neural network (CorrNet), that explicitly maximizes correlation among the views when projected to the common subspace and shows that the representations learned using it perform better than the ones learned using other state-of-the-art approaches. Expand
Efficient Computation of the Shapley Value for Game-Theoretic Network Centrality
This paper develops exact analytical formulae for Shapley value-based centrality in both weighted and unweighted networks and develops efficient and exact algorithms based on them and demonstrates that they deliver significant speedups over the Monte Carlo approach. Expand
An algebraic approach to abstraction in reinforcement learning
To operate effectively in complex environments learning agents ignore irrelevant details. Stated in general terms this is a very difficult problem. Much of the work in this field is specialized toExpand
Learning to Repeat: Fine Grained Action Repetition for Deep Reinforcement Learning
A novel framework, Fine Grained Action Repetition (FiGAR), which enables the agent to decide the action as well as the time scale of repeating it and can be used for improving any Deep Reinforcement Learning algorithm which maintains an explicit policy estimate by enabling temporal abstractions in the action space. Expand
Accurate mobile robot localization in indoor environments using bluetooth
Novel approaches for obtaining distance estimates and trilateration that overcome the hitherto known limitations of using bluetooth for localization are introduced that have the potential of being scaled to multi-agent scenarios. Expand
Efficient Computation of the Shapley Value for Centrality in Networks
This paper presents the first such study of the Shapley Value for network centrality, a measure of node centrality of paramount significance in many real-world application domains including social and organisational networks, biological networks, communication networks and the internet. Expand
Approximate Homomorphisms : A framework for non-exact minimization in Markov Decision Processes
To operate effectively in complex environments learning agents require the ability to selectively ignore irrelevant details and form useful abstractions. In earlier work we explored in detail whatExpand