Toby Kreiman

I'm a PhD student at UC Berkeley in BAIR where I am advised by Aditi Krishnapriyan. I studied Physics and Computer Science at Columbia University during my undergraduate degree, where I was fortunate to be advised by Matei Ciocarlie.

I am interested in the intersection of Physics and Machine Learning. I am currently working on developing physics-inspired machine learning algorithms for molecular dynamics. I'm also interested in Reinforcement Learning; I have done previous work on developing algorithms for co-design of hardware and policy for robotic manipulation.

Email / CV / Scholar / Github

Research

Education

University of California, Berkeley (Aug 2023 - Present)

Ph.D. student in Computer Science

Columbia University (Aug 2019 - May 2023)

B.A. in Computer Science and Physics

Work Experience

Google (Jun 2022 - Aug 2022)

Machine Learning Engineer for Youtube Autoplay.

Meta (Jun 2021 - Aug 2021)

Machine Learning Engineer on podcast recommendation team.

Supportiv (Jun 2020 - Aug 2020)

Machine Learning Engineer.

	Understanding and Mitigating Distribution Shifts For Machine Learning Force Fields Tobias Kreiman, Aditi Krishnapriyan Digital Discovery, 2026 project page / code / paper
	Transformers Discover Molecular Structure Without Graph Priors Tobias Kreiman, Yutong Bai, Fadi Atieh, Elizabeth Weaver, Eric Qu, Aditi Krishnapriyan Preprint, 2025 project page / code / paper
	Action-Minimization Meets Generative Modeling: Efficient Transition Path Sampling with the Onsager-Machlup Functional Sanjeev Raja, Martin Šípka, Michael Psenka, Tobias Kreiman, Michal Pavelka, Aditi Krishnapriyan ICML, 2025 code / paper
	MLIP Arena: Advancing Fairness and Transparency in Machine Learning Interatomic Potentials through an Open and Accessible Benchmark Platform Yuan Chiang, Tobias Kreiman, Elizabeth Weaver, Ishan Amin, Matthew Kuner, Christine Zhang, Aaron Kaplan, Daryl Chrzan, Samuel M Blau, Aditi Krishnapriyan Mark Asta, AI4MAT Workshop Spotlight @ICLR, NeurIPS Spotlight, 2025 project page / code / paper
	Foundation Policies with Hilbert Representations Seohong Park, Tobias Kreiman, Sergey Levine, ICML, 2024 project page / code / paper
	Octo: An Open-Source Generalist Robot Policy Dibya Ghosh, Homer Walke, Karl Pertsch, Kevin Black, Oier Mees, Sudeep Dasari, Joey Hejna, Charles Xu, Jianlan Luo, Tobias Kreiman, You Liang Tan, Dorsa Sadigh, Chelsea Finn, Sergey Levine, RSS*, 2024 project page / code / paper