I am a PhD candidate in the Department of Biomedical Engineering at Johns Hopkins University. I broadly work on {statistical, deep} learning, with an aspiration of reducing the gap between machine and natural intelligence. My doctoral research focuses on learning under non-stationary distributions and distribution shifts, with applications in large language models, computer vision, and biomedical data science.
-
Johns Hopkins University
- Baltimore, MD
-
16:21
(UTC -12:00) - https://laknath1996.github.io/
- @AshwindeSilva1
- in/ashwin-de-silva-6852b14b
Pinned Loading
-
neurodata/prolearn
neurodata/prolearn PublicProspective Learning: Learning for a Dynamic Future (NeurIPS 2024)
Python 4
-
neurodata/value-of-ood-data
neurodata/value-of-ood-data PublicThe value of out-of-distribution data (ICML 2023)
Python 9
-
gen-models
gen-models PublicCourse project for EN.553.741 Machine Learning II at JHU. Implements Wasserstein GAN, variational autoencoder and denoising diffusion probabilistic models.
Python
-
semi-supervised-learning
semi-supervised-learning PublicRepository for the EN.553.738 High-Dimensional Approximation, Probability and Statistical Learning group project on graph-based semi-supervised learning
MATLAB 1
-
DeepPhaseUnwrap
DeepPhaseUnwrap PublicThis repository Introduces a joint convolutional and spatial quad-directional LSTM (SQD-LSTM) network for phase unwrapping in 2D images (ICASSP 2021)
-
sEMG-Hand-Gesture-Recognition
sEMG-Hand-Gesture-Recognition PublicThe source code for the real-time hand gesture recognition algorithm based on Temporal Muscle Activation maps of multi-channel surface electromyography (sEMG) signals (ICASSP 2021)
If the problem persists, check the GitHub status page or contact support.