Updates
2024-12 We are organizing a workshop on weight space learning at ICLR 2025! Details on weight-space-learning.github.io
2024-12 I will be attending NeurIPS in person this year! Happy to learn about symmetry, LLMs, and everything. We will present two extended abstracts at the workshops:
• Symmetry Discovery in Neural Network Parameter Spaces at UniReps
• Emergence of Hierarchical Emotion Representations in Large Language Models at SciForDL.
2024-12 Honored to be selected as an NVIDIA Graduate Fellowship Finalist.
2024-08 Serving as the Student Program and Funding Chair of the WiML workshop at NeurIPS 2024.
2024-06 Interning this summer at NTT Research in Cambridge, MA.
2024-04 Contributed talk on Symmetries, flat minima, and the conserved quantities of gradient flow at SOCAMS (Southern California Applied Mathematics Symposium).
2024-03 Our paper Understanding the Difficulty of Solving Cauchy Problems with PINNs was accepted by The 6th Annual Learning for Dynamics and Control Conference (L4DC 2024).
2024-01 Our paper Improving Convergence and Generalization Using Parameter Symmetries was accepted by ICLR 2024 for an oral presentation.
2023-10 Two extended abstracts were accepted by NeurIPS workshops:
• Improving Convergence and Generalization Using Parameter Symmetries at NeurReps
• Understanding Mode Connectivity via Parameter Space Symmetry at UniReps.
2023-09 Our paper DYffusion: A Dynamics-informed Diffusion Model for Spatiotemporal Forecasting was accepted by NeurIPS 2023.
2023-07 Talk on Symmetries, flat minima, and the conserved quantities of gradient flow at Boston Computation Club.
2023-06 Honored to be selected to receive the DeepMind Fellowship.
2023-01 Our paper Symmetries, flat minima, and the conserved quantities of gradient flow was accepted by ICLR 2023.
2022-11 Contributed talk on Symmetry Teleportation at Conference on the Mathematical Theory of Deep Neural Networks (DeepMath 2022)
2022-09 Our paper Symmetry Teleportation for Accelerated Optimization was accepted by NeurIPS 2022.
2022-06 Interning this summer at the MIT-IBM Watson AI Lab!
2022-05 Our paper LIMO: Latent Inceptionism for Targeted Molecule Generation was accepted by ICML 2022.
2022-04 Our paper Concentric Spherical Neural Network for 3D Representation Learning was accepted by IJCNN 2022.
2021-09 Joined the Spatiotemporal Machine Learning Lab and started PhD at UC San Diego!