Maximum Entropy Model Correction in Reinforcement Learning
A. Rakhsha, M. Kemertas, M. Ghavamzadeh, A.M. Farahmand
To appear in International Conference on Learning Representations (ICLR), 2024
Publications
Efficient and Accurate Optimal Transport with Mirror Descent and Conjugate Gradients
M. Kemertas, A. Jepson, A.M. Farahmand
Preprint, 2023
Approximate Policy Iteration with Bisimulation Metrics
M. Kemertas, A. Jepson
Published in Transactions on Machine Learning Research (TMLR), 2022
CrispSearch: Low-latency On-device Language-based Image Retrieval
Z. Hu*, L. Xiao*, M. Kemertas*, C. Phillips, I. Mohomed, A. Fazly (* denotes equal contribution)
Published in ACM Multimedia Systems Conference, 2022
Towards Robust Bisimulation Metric Learning
M. Kemertas, T.T. Aumentado-Armstrong (equal contribution)
Published in Neural Information Processing Systems (NeurIPS), 2021
Dependency Parsing with Structure Preserving Embeddings
Á. Kádár, L. Xiao, M. Kemertas, F. Fancellu, A. Jepson and A. Fazly
Published in Conference of the European Chapter of the Association for Computational Linguistics (EACL), 2021
RankMI: A Mutual Information Maximizing Ranking Loss
M. Kemertas, L. Pishdad, K.G. Derpanis, A. Fazly
Published in IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020
Dynamic Scheduling of MPI-based Distributed Deep Learning Training Jobs
T. Capes, V. Raheja, M. Kemertas, I. Mohomed
Published in MLSys Workshop at Neural Information Processing Systems (NeurIPS), 2018