1 code implementation • 2 Mar 2023 • Siddhant Haldar, Jyothish Pari, Anant Rai, Lerrel Pinto
Given a weak base-policy trained by offline imitation of demonstrations, FISH computes rewards that correspond to the "match" between the robot's behavior and the demonstrations.
no code implementations • 30 Jun 2022 • Siddhant Haldar, Vaibhav Mathur, Denis Yarats, Lerrel Pinto
Our experiments on 20 visual control tasks across the DeepMind Control Suite, the OpenAI Robotics Suite, and the Meta-World Benchmark demonstrate an average of 7. 8X faster imitation to reach 90% of expert performance compared to prior state-of-the-art methods.
1 code implementation • 11 Jul 2020 • Avisek Lahiri, Sourav Bairagya, Sutanu Bera, Siddhant Haldar, Prabir Kumar Biswas
We also present and analyse our results highlighting the drawbacks of applying depthwise separable convolutional kernel (a popular method for efficient classification network) for sub-pixel convolution based upsampling (a popular upsampling strategy for low-level vision applications).