My research mainly focuses on Reinforcement Learning (RL) under the supervision of Emma Brunskill in Stanford AI for Human Impact, Stanford Artificial Intelligence Lab and Statistical Machine Learning Group.

More specifically, I study the following topics:

• Robust and Safe Reinforcement Learning
• Off-policy policy evaluation and optimization
• Application of RL to health care

Conference Publications

Off-policy Policy Evaluation For Sequential Decisions Under Unobserved Confounding [arXiv]

Thirty-fourth Conference on Neural Information Processing Systems, NeurIPS 2020

Hongseok Namkoong ^* , Ramtin Keramati ^* , Steve Yadlowsky ^* , Emma Brunskill

Being Optimistic to Be Conservative: Quickly Learning a CVaR Policy [arXiv]

The Thirty-Fourth AAAI Conference on Artificial Intelligence, AAAI 2020, New York, USA

Ramtin Keramati, Chris Dann, Alex Tamkin, Emma Brunskill

Value Driven Representations for Reinforcement Learning [arXiv]

User Modelling, Adaptation and Personalization (UMAP) 2019, (23% acceptance rate)

Ramtin Keramati, Emma Brunskill

Identification of Subgroups With Similar Benefits in Off-Policy Policy Evaluation

Under submission: Thirty-eighth International Conference on Machine Learning, ICML 2021

Ramtin Keramati, Omer Gottesman, Finale Doshi-Velez, Emma Brunskill

Journal Publications

Significant contribution of small icebergs to the freshwater budget in Greenland fjords[link]

Communications earth & environment, Nature Publishing Group

Soroush Rezvanbehbahani, Leigh A Stearns, Ramtin Keramati, Siddharth Shankar, CJ van der Veen

Dynamics of the nanoneedle probe in trolling mode AFM [link]

Journal of Nanotechnology

Ahmad Abdi, Hossein Nejat Pishkenari, Ramtin Keramati, Majid Minary-Jolandan

Molecular dynamics modeling of a nanomaterials-water surface interaction [link]

Journal of Applied Physics

Hossein Nejat Pishkenari, Ramtin Keramati, Ahmad Abdi, Majid Minary-Jolandan

Workshop Publications

Off-policy Policy Evaluation Under Unobserved Confounding [link, presentation]

ICLR 2020: Causal Learning For Decision Making

Hongseok Namkoong ^* , Ramtin Keramati ^* , Steve Yadlowsky ^* , Emma Brunskill

Distributionally-Aware Exploration for CVaR Bandits [pdf]

NeurIPS 2019 : Safety and Robustness in Decision Making workshop

Alext Tamkin, Ramtin Keramati, Chris Dann, Emma Brunskill

Strategic Object Oriented Reinforcement Learning [arXiv]

ICML 2018: Exploration in Reinforcement Learning workshop

Ramtin Keramati, Jay Whang, Patrick Cho, Emma Brunskill

Hierarchy-Driven Exploration in Reinforcement Learning [pdf]

ICML 2018: Exploration in Reiforcement Learning workshop

Evan Liu, Ramtin Keramati, Kelvin Guu, Sudarshan Seshadri, Panupong Pasupat, Percy Liang and Emma Brunskill

Teaching

• Award: Centennial TA 2018
• Head TA of Reinforcement Learning (CS234), Prof. Emma Brunskill
• Teaching assistant of Reinforcement Learning (CS234), Prof. Emma Brunskill
• Head TA of Deep Learning (CS230), Prof. Andrew Ng, Kian Katanforoosh
• Teaching assistant of Deep Learning (CS230), Prof. Andrew Ng, Kian Katanforoosh
• Teaching assistant of Linear Algebra with Application (CME200), Prof. Iaccarino, G.
• Teaching assistant of Numerical Methods for Engineering (CME206), Prof. Marsden, A.