Shruti Mishra

I am a multidisciplinary scientist with a focus on reinforcement learning and continuum mechanics. I work as a Research Scientist in reinforcement learning at Sony AI, developing methods for PlayStation games and open-source domains. I earned my PhD in Applied Mathematics from Harvard University, where I worked on problems in continuum mechanics, computation and machine learning. I have spent some time learning and doing research at DeepMind, on reinforcement learning in continuous control environments, and at Mila, on the use of deep learning to forecast natural convection.

Education

PhD, Applied Mathematics, Harvard University (2021)
SM, Applied Mathematics, Harvard University (2017)
MEng, Mechanical Engineering, Imperial College London (2014)

Publications

Policy composition via multi-objective reinforcement learning
S. Mishra, A. Anand, J. Hoffmann, N. Heess, M. Riedmiller, A. Abdolmaleki, D. Precup
Conference on Robot Learning, LocoLearn workshop (2024) [article] [poster]

Computing the viscous effect in early-time drop impact dynamics
S. Mishra, S.M. Rubinstein, C.H. Rycroft
Journal of Fluid Mechanics (2022) [article] [code]

On Multi-objective Policy Optimization as a Tool for Reinforcement Learning
A. Abdolmaleki, S.H. Huang, G. Vezzani, B. Shahriari, J.T. Springenberg, S. Mishra, D. Tirumala, A. Byravan, K. Bousmalis, A. György, C. Szepesvari, R. Hadsell, N. Heess, M. Riedmiller (2021) [article]

Coordinated crawling via reinforcement learning
S. Mishra, W.M. van Rees, L. Mahadevan
Journal of the Royal Society Interface (2020) [article]

Augmenting learning using symmetry in a biologically-inspired domain
S. Mishra, A. Abdolmaleki, A. Guez, P. Trochim, D. Precup
Neural Information Processing Systems, NewInML workshop (2019) [article]

Machine learning in a data-limited regime: Augmenting experiments with synthetic data uncovers order in crumpled sheets
J. Hoffmann, Y. Bar-Sinai, L.M. Lee, J. Andrejevic, S. Mishra, S.M. Rubinstein, C.H. Rycroft
Science Advances (2019) [article]

Thermal design of the FETS chopper beam dump
S. Mishra, M. Aslaninejad, P. Savage, P.A. Posocco, J.K. Pozimski, A.P. Letchford
Proceedings of the Particle Accelerator Conference (2013) [article]

Professional Activities

Reviewing

Conference on Neural Information Processing Systems
Physical Review Letters
Reinforcement Learning Conference
AAAI Conference on Artificial Intelligence, workshops
International Conference on Machine Learning, workshops

Organising

Workshop at RLDM 2025, Saving the Phenomena of Minds
RL Social at ICML 2021
Workshop at NeurIPS 2020, NewInML

Teaching and Mentoring

Teaching Fellow, Harvard University: CS249r Autonomous Systems, AM205 Advanced Scientific Computing, AM115 Mathematical Modeling, ES220 Fluid Mechanics
Teaching Assistant, Imperial College: ME1 Mathematics
Alumni Mentor, Imperial College

Panels

Harvard alumnae panel, 2024
Imperial offer holders panel, 2022

Awards

G-Research Grant for students and postdocs in quantitative fields (2021)   [description]
Summer Research Grant, Harvard University (2021)
Best Oral Presentation Award, NewInML workshop, NeurIPS conference (2019)
Young Scientist Travel Award, Harvard Brain Initiative (2019)   [description]
Certificate of Distinction in Teaching, Harvard University (2019, 2017)   [description]
Bramwell Medal, Imperial College (2014)   [description] [photo]
Henry Ford II Scholar Award in Mechanical Engineering, Imperial College (2014)   [description]
Santander Final Year Scholarship, Imperial College (2013)  
Frank Turner Wilson Prize, Imperial College (2013)   [description]
Sir Bruce White Laboratory Prize in Mechanical Engineering, Imperial College (2012)   [description]
Frank Turner Wilson Prize, Imperial College (2011)   [description]
Faculty of Engineering Dean's List, Imperial College (2014, 2013, 2012, 2011)  
Gold Prize and Best in school, UK Maths Challenge (2010)  

Miscellaneous

Photography [flickr]   ·   Gardening

News

June 2025: Co-organising a workshop at RLDM
May 2025: Invited talk at the AI4Good Lab on Applications of reinforcement learning in locomotion