Decentralized Autonomous Networks for Cooperative Estimation

Ryan Goldhahn | 20-SI-005

Executive Summary

We are developing algorithms, simulation tools, and hardware implementations for intelligent autonomous sensor networks that will enable them to extract information from data and adapt to uncertain environments and potential adversaries without a centralized command-and-control agent. These are critical advances for more intelligent, more robust sensor networks for potential use in a variety of national security missions.

Publications, Presentations, and Patents

Rusu, Edward, and Ruben Glatt. "Abmarl: Connecting Agent-Based Simulations with Multi-Agent Reinforcement Learning," Journal of Open Source Software 6, no. 64 (2021): 3424.

Cadena, Jose, Priyadip Ray, Hao Chen, Braden Soper, Deepak Rajan, Anton Yen, and Ryan Goldhahn. "Stochastic Gradient-Based Distributed Bayesian Estimation in Cooperative Sensor Networks," IEEE Transactions on Signal Processing 69 (2021): 1713-1724.

Cheng, Hao, Kaidi Xu, Chenan Wang, Xue Lin, Bhavya Kailkhura, and Ryan Goldhahn. "Mixture of Robust Experts (MoRE): A Flexible Defense Against Multiple Perturbations." Robust and Reliable ML in the Real World Workshop at ICLR. 2021.

Cheng, Hao, Kaidi Xu, Chenan Wang, Xue Lin, Bhavya Kailkhura, and Ryan Goldhahn. "More or Less (MoL): Defending against Multiple Perturbation Attacks on Deep Neural Networks through Model Ensemble and Compression." IEEE Winter Conference on Applications of Computer Vision. Under review, 2021.

Dawson, William A., Ruben Glatt, Edward Rusu, Braden C. Soper, and Ryan A. Goldhahn. "Hybrid Information-driven Multi-agent Reinforcement Learning." Challenges and Opportunities for Multi-Agent Reinforcement Learning (COMARL) AAAI Conference. 2021.

Chen, Cheng, Bhavya Kailkhura, Ryan Goldhahn, and Yi Zhou. "Certifiably-Robust Federated Adversarial Learning via Randomized Smoothing." IEEE Mobile Ad Hoc and Smart Systems (MASS) Conference. Accepted, 2021.

Cadena, Jose, Priaydip Ray, and Ryan Goldhahn. “Decentralized Black-Box Variational Inference for Bayesian Learning on Sensor Networks.” Asilomar Conference on Signals, Systems, and Computers. Accepted, 2021.

Goldhahn, Ryan Alan, Priyadip Ray, Braden C. Soper, Hao Chen, and Deepak Rajan. "Identification of a characteristic of a physical system based on collaborative sensor networks." U.S. Patent Application 17/165,714, filed August 5, 2021.