no code implementations • 28 Dec 2022 • Riashat Islam, Hongyu Zang, Manan Tomar, Aniket Didolkar, Md Mofijul Islam, Samin Yeasar Arnob, Tariq Iqbal, Xin Li, Anirudh Goyal, Nicolas Heess, Alex Lamb
Several self-supervised representation learning methods have been proposed for reinforcement learning (RL) with rich observations.
no code implementations • AAAI 2022 • Md Mofijul Islam, Tariq Iqbal
However, it is challenging to extract robust multimodal representations due to the heterogeneous characteristics of data from multimodal sensors and disparate human activities, especially in the presence of noisy and misaligned sensor data.
Ranked #1 on Multimodal Activity Recognition on MMAct
no code implementations • 1 Jul 2021 • Mohammad Samin Yasar, Tariq Iqbal
These variables make it challenging for learning algorithms to obtain a general representation that is robust to the diverse spatio-temporal patterns of human motion.
no code implementations • IEEE ROBOTICS AND AUTOMATION LETTERS 2021 • Md Mofijul Islam, Tariq Iqbal
Finally, the experimental results with noisy sensor data indicate that Multi-GAT consistently outperforms all the evaluated baselines.
Ranked #2 on Multimodal Activity Recognition on MMAct
Human Activity Recognition Multimodal Activity Recognition +1
no code implementations • International Conference on Intelligent Robots and Systems (IROS) 2020 • Md Mofijul Islam, Tariq Iqbal
We develop a novel multimodal attention mechanism for disentangling and fusing the salient unimodal features to compute the multimodal features in the upper layer.
no code implementations • 4 May 2016 • Tariq Iqbal, Samantha Rack, Laurel D. Riek
We compared the results of our anticipation method to move the robot with another method which did not rely on high-level group behavior, and found our method performed better both in terms of more closely synchronizing the robot's motion to the team, and also exhibiting more contingent and fluent motion.