PyTorch: I’m Fast, JAX: You Call That Fast?
A recipe to train Object Detection Transformers (really) fast.
Posts by Collection
portfolio
A GPU Poor’s Multi-node Training Cluster
Bag-of-tricks for multi-node training using plain Ethernet.
publications
XLDA: Linear Discriminant Analysis for Incremental Extreme Classification
Published in International Conference on Machine Learning (ICML), PAC-Bayes Workshop, 2023
An O(1) on-device continual learning classifier.
Recommended citation: Shah, K., Veerendranath, V., Hebbar, A. and Bhat, R., 2023. XLDA: Linear Discriminant Analysis for Scaling Continual Learning to Extreme Classification at the Edge. arXiv preprint arXiv:2307.11317.
Download Paper
talks
Interpretability in Deep Learning
Published:
This talk focuses on the importance and methods of making machine learning models, particularly deep learning models, interpretable and understandable. Computation and search has eliminated the need for domain-specific feature engineering. However, this has a caveat - deep models often behave like black boxes, and are prone to issues of trust, transparency, and safety, especially in mission-critical applications.
Federated Learning for Healthcare
Published:
The aim of this tutorial is to facilitate education on how to perform Federated Learning on both simulated and real world studies from software-based privacy-preserving techniques (e.g. DP), to hardware-based trusted execution environments (TEEs).