Share This Author
Belief Propagation Neural Networks
By training BPNN-D, a learned iterative operator that provably maintains many of the desirable properties of BP for any choice of the parameters, BPNNs learns to perform the task better than the original BP: it converges 1.7x faster on Ising models while providing tighter bounds.
IQ-Learn: Inverse soft-Q Learning for Imitation
- Divyansh Garg, Shuvam Chakraborty, Chris Cundy, Jiaming Song, S. Ermon
- Computer ScienceNeurIPS
- 23 June 2021
This work introduces a method for dynamics-aware IL which avoids adversarial training by learning a single Q-function, implicitly representing both reward and policy, and obtains state-of-the-art results, outperforming existing methods both in the number of required environment interactions and scalability in high-dimensional spaces, often by more than 3x.
Efficient Conditional Pre-training for Transfer Learning
- Shuvam Chakraborty, B. Uzkent, Kumar Ayush, K. Tanmay, Evan Sheehan, S. Ermon
- Computer ScienceArXiv
- 20 November 2020
This work proposes efficient target dataset conditioned filtering methods to remove less relevant samples from the pre-training dataset and discovers that lowering image resolutions in the pre -training step offers a great trade-off between cost and performance.
Learning from Peers at the Wireless Edge
- Shuvam Chakraborty, Hesham Mohammed, D. Saha
- Computer ScienceInternational Conference on COMmunication Systems…
- 1 January 2020
This paper introduces a peer to peer Federated Learning model, where a local model is trained based on the sensing results of each node and shared among its peers to create a global model, and generates wireless channel access data which is used to train the local models.
Domain Knowledge aided Neural Network for Wireless Channel Estimation
- Shuvam Chakraborty, D. Saha
- Computer Science, BusinessIEEE Global Communications Conference (GLOBECOM)
- 1 December 2021
This work proposes a NN model, where the structure and parameters are derived from the domain knowledge of wireless signal and channel characteristics, and achieves ∼10 dB improvement over Least Square channel estimation.
Because It ' s the Cup : Predicting the Stanley Cup
The Stanley Cup playoffs have long been known for their drama: most games are close, upsets are common, and teams not considered one of the best can win the cup. However, current predictions are…