Samuel A. Barnett

Bio

I am a 5th year PhD student in AI and Machine Learning at Princeton University, advised by Profs. Ryan Adams and Tom Griffiths. My work focuses on reinforcement learning, social cognition, and multi-agent systems. Previously, I interned at Microsoft Research New York, where I worked on projects in cognitive neuroscience with Dr Ida Momennejad.

Papers

• Barnett, S. A., Aduol, J., Guerra, A., Gumussoy, S., and Adams, R. P. (in submission). Toeplitz posterior approximation for multivariate time series modeling with Gaussian processes.
• Draft available upon request.
• Barnett, S. A., Griffiths, T. L., & Hawkins, R. D. (2022). A pragmatic account of the weak evidence effect. Open Mind, 6, pp. 169-182..
  • Paper
  • Code
• Barnett, S. A., & Momennejad, I. (2022). PARSR: Priority-Adjusted Replay for Successor Representations. 5th Multidisciplinary Conference on Reinforcement Learning and Decision Making (RLDM 2022) (pp. 80-85). RLDM.
  • Paper
  • Code
• Juliani, A., Barnett, S. A., Davis, B., Sereno, M., & Momennejad, I. (2022). Neuro-Nav: A Library for Neurally-Plausible Reinforcement Learning. 5th Multidisciplinary Conference on Reinforcement Learning and Decision Making (RLDM 2022) (pp. 290-294). RLDM.
  • Paper
  • Code
• Barnett, S. A. (2018). Convergence problems with generative adversarial networks (GANs). arXiv preprint arXiv:1806.11382.
  • Paper
• Barnett, S. A. (2018). What decision theory provides the best procedure for identifying the best action available to a given artificially intelligent system? [Unpublished MMathPhil thesis]. University of Oxford.
  • Paper

Contact

samuelab at cs dot princeton dot edu

Links

CV

GitHub

Google Scholar