Ashwin De Silva
Applied Scientist at Amazon
Hello there! I’m an Applied Scientist at Amazon. I completed my PhD in Biomedical Engineering at Johns Hopkins University in May 2026, where I was advised by Profs. Joshua Vogelstein, Pratik Chaudhari, and Carey Priebe. My doctoral research focused on learning from non-stationary data and culminated in a mathematical framework called prospective learning. This work connects closely to ongoing efforts in continual learning and out-of-distribution (OOD) generalization.
More broadly, I’m driven by two goals: advancing AI by designing new models and algorithms with strong theoretical guarantees, and building efficient, robust learning systems that hold up in a dynamic world. My interests span generative modeling, reinforcement learning, continual learning, learning theory, and computer vision.
I also hold a Master’s in Applied Mathematics and Statistics from Johns Hopkins (2024) and a Bachelor’s in Biomedical Engineering from the University of Moratuwa, Sri Lanka (2020).
Outside work, you’ll find me spending time with my wife, Malsha, playing the piano, or out hiking and swimming. I’m also an avid history, philosophy and astronomy enthusiast!
Recent news
- Jan 2026 Optimal control of the future via prospective learning with control accepted to L4DC 2026.
- Oct 2025 Prospective Learning in Retrospect won the student best paper award at AGI 2025.
- Jul 2025 Presented a poster at the NeuroAI conference (Allen Institute, Seattle).
- Jun 2025 Prospective Learning in Retrospect accepted to AGI 2025 (oral).
- Jun 2025 Simple Calibration via Geodesic Kernels accepted to TMLR.
Selected awards
- 2025 Best Student Paper Award, AGI 2025
- 2025 Alpha Eta Mu Beta membership
- 2024 MINDS Fellowship, Johns Hopkins University
- 2023 Student Spotlight, Johns Hopkins School of Medicine
- 2022 Best Short Paper Award, ECCV OOD Workshop
- 2021 Second Runner-Up, IEEE VIP Cup at ICIP
- 2020 Gold Medal, University of Moratuwa
- 2020 Prof. Pathuwathawithana Memorial Prize
Selected publications
-
Prospective Learning: Learning for a Dynamic Future
Ashwin De Silva, Rahul Ramesh, Rubing Yang, Siyu Yu, Pratik Chaudhari, Joshua T. Vogelstein
NeurIPS 2025 [pdf] -
Simple Calibration via Geodesic Kernels
Jayanta Dey, Haoyin Xu, Ashwin De Silva, Joshua T. Vogelstein
TMLR 2025 [pdf] -
The Value of Out-of-Distribution Data
Ashwin De Silva, Rahul Ramesh, Carey E. Priebe, Pratik Chaudhari, Joshua T. Vogelstein
ICML 2023 [pdf] -
A Joint Convolutional and Spatial Quad-Directional LSTM Network for Phase Unwrapping
Malsha V. Perera, Ashwin De Silva
ICASSP 2021 [pdf] -
Real-time Hand Gesture Recognition Using Temporal Muscle Activation Maps of Multi-channel sEMG Signals
Ashwin De Silva, Malsha V. Perera, Kithmin Wickramasinghe, Asma M. Naim, Thilina Dulantha Lalitharatne, Simon L. Kappel
ICASSP 2020 [pdf]
Teaching
- Fall 2025 EN.580.697 Biomedical Data Design (TA), Johns Hopkins University
- Fall 2020 EN 1060 Signals and Systems (Junior Lecturer), University of Moratuwa
- Spring 2020 EN 3030 Circuits and Systems Design (Junior Lecturer), University of Moratuwa
- Spring 2020 BM 2101 Analysis of Physiological Systems (Junior Lecturer), University of Moratuwa