Saeid Amiri

I am research scientist at JP Morgan AI research working at the intersection of AI and Finance. Previously, I was a student at AIR Lab, SUNY Binghamton, working in the AIR lab focusing on knowledge-based robotics sequential decision making algorithms under the supervision of Shiqi Zhang. More specifically, I am interested in leveraging commonsense knowledge in POMDPs, Reinforcement Learning, and combinatorial optimization.

Email  /  Google Scholar  /  Github

I'm interested in AI symbolic planning under uncertainty and Reinforcement Learning. Much of my research is about developing sequential decision making frameworks under partial observability. Representative papers are highlighted.

Using Surrogates to speedup MCTS in combinatorial optimization

Vision-language models (VLMs) have been applied to robot task planning problems

Using supervised learning and automated reasoning to guide a POMDP-based symbolic planner

A metric for evaluating fraud detection system performance.

Using action knowledge to guide the model-based RL.

Using supervised learning and automated reasoning to guide a POMDP-based symbolic planner

A robotic dialogue system that simultaneously identifies human intention and augments its knowledge base.

Leveraging Multi-Sensory data to guide a robot to identify object properties.

Combinatorial Optimization

Vision-based grasping using deep reinforcement learning

Teaching First-year students AI and robotics and mentor them in their research projects.

Template Source