Research Interests: Reinforcement Learning (RL), Transfer Learning in RL, Model-Based RL, Natural Language Processing, Probing Neural Language Models, Generative Dialogue models.

Publications (reports, proceedings and pre-prints)

2022

  • Paul-Aymeric McRae, Prasanna Parthasarathi, Mahmoud Assran and Sarath Chandar (2022). Memory Augmented Optimizers for Deep Learning. ICLR.

  • Louis Clouâtré, Prasanna Parthasarathi, Amal Zouaq and Sarath Chandar (2022). Local Structure Matters Most: Perturbation Study in NLU. ACL [Findings].

2021

  • Prasanna Parthasarathi, Koustuv Sinha, Joelle Pineau and Adina Williams (2021). Sometimes We Want Translationese.EMNLP [Findings].

  • Koustuv Sinha, Prasanna Parthasarathi, Joelle Pineau and Adina Williams (2021). Unnatural Language Inference. ACL[Outstanding Paper award].

  • Prasanna Parthasarathi, Joelle Pineau, Sarath Chandar (2021). Do Encoder Representations of Generative Dialogue Models Encode Sufficient Information about the Task ? SIGDial.

  • Prasanna Parthasarathi, Mohamed Abdelsalam, Joelle Pineau, Sarath Chandar (2021). A Brief Study on the Effects of Training Generative Dialogue Models with a Semantic loss. SIGDial.

2020

  • Arvind Neelakantan, Semih Yavuz, Sharan Narang, Vishaal Prasad, Prasanna Parthasarathi, Ben Goodrich, Daniel Duckworth, Chinnadhurai Sankar, Xifeng Yan (2020). Neural Assistant: Joint Action Prediction, Response Generation and Latent Knowledge Reasoning, arXiv.

  • Prasanna Parthasarathi, Arvind Neelakantan, Sharan Narang (2020). On Task-Level Dialogue Composition of Generative Transformer Model. EMNLP-Workshop.

  • Koustuv Sinha, Prasanna Parthasarathi, Jasmine Wang, Ryan Lowe, William L. Hamilton, Joelle Pineau (2020). Learning an Unreferenced Metric for Online Dialogue Evaluation. ACL.

2018

  • Nicolas Gontier, Koustuv Sinha, Peter Henderson, Iulian Serban, Michael Noseworthy, Prasanna Parthasarathi, Joelle Pineau (2018)The RLLChatbot: a solution to the ConvAI challenge , arXiv.

  • Prasanna Parthasarathi, and Joelle Pineau (2018) Extending Neural Generative Conversational Model using External Knowledge Sources ,EMNLP.

  • Nicolas Angelard-Gontier, Joshua Romoff, Prasanna Parthasarathi Variational Encoder-Decoder trained to read math operations and produce the result as an MNIST image, here.

2017

  • Prasanna Parthasarathi (2017) Understanding Exploration Strategies in Model Based Reinforcement Learning. Masters' Thesis, Indian Institute of Technology Madras. here.

  • Hoai Phuoc Truong, Prasanna Parthasarathi, and Joelle Pineau (2017) MACA: A Modular Architecture for Conversational Agents, SIGDial.

  • Aravind Lakshminarayanan, Janarthanan Rajendran, Mitesh M. Khapra,Prasanna Parthasarathi, and Balaraman Ravindran (2017) A2T: Attend Adapt and Transfer: Attentive Deep Architecture for Adaptive Transfer from multiple sources, ICLR.

2015

  • Janarthanan Rajendran, Prasanna Parthasarathi, Balaraman Ravindran, and Mitesh M. Khapra (2015)ADAAPT: A Deep Architecture for Adaptive Policy Transfer from Multiple Sources . NeurIPS-Workshop.

  • Vaishnavh Nagarajan, Prasanna Parthasarathi, Sutanu Chakraborti (2015) Readability Estimation of Documents . arXiv,here.

  • Prasanna Parthasarathi, Sarath Chandar, and Ravindran Balaraman (2015) Thompson Sampling with Adaptive Exploration Bonus for Near-Optimal Policy Learning. RLDM.

2014

  • Prasanna Parthasarathi, Sarath Chandar, and Balaraman Ravindran (2014) iBayes: A Thompson Sampling Approach to ReinforcementLearning with Instructions. NeurIPS-Workshop.

  • Prasanna Parthasarathi, Sethuraman Sundararaman, Manigandan N., Ramkumar Kannan (2014) Fuzzy-Based Obstacle Navigation and Effective Path Selection with Laser Range Finder. J. Applied Sci.