
Harsha Kokel is a Research Scientist at IBM Research. She obtained her Ph.D. at The University of Texas at Dallas under the guidance of Prof. Sriraam Natarajan. She is interested in knowledge-guided learning for structured, relational domains. Her research focuses on leveraging domain-specific expert knowledge to improve generalization in various AI/ML approaches. She is interested in sequential decision-making problems in structured domains and explores the combination of planning and reinforcement learning. Recently, she has also ventured into tabular representation learning for data discovery tasks.
Her research has been published in the following conferences: AAAI, ICAPS, IJCAI, NeurIPS, and ACS. Her work has been presented at various workshops at AAAI, NeurIPS, IJCNN, ICAPS, IJCAI and RLDM. She has reviewed papers for various journals, conferences, and workshops, including AAAI, IJCAI, NeurIPS, SDM, and DMKD. She serves as an assistant electronic publishing editor for JAIR. She co-organized the workshop on Bridging the Gap Between AI Planning and Reinforcement Learning (PRL) at IJCAI and ICAPS 2023.