Publications

Refine Results

(Filters Applied) Clear All

Towards a distributed framework for multi-agent reinforcement learning research

Summary

Some of the most important publications in deep reinforcement learning over the last few years have been fueled by access to massive amounts of computation through large scale distributed systems. The success of these approaches in achieving human-expert level performance on several complex video-game environments has motivated further exploration into the limits of these approaches as computation increases. In this paper, we present a distributed RL training framework designed for super computing infrastructures such as the MIT SuperCloud. We review a collection of challenging learning environments—such as Google Research Football, StarCraft II, and Multi-Agent Mujoco— which are at the frontier of reinforcement learning research. We provide results on these environments that illustrate the current state of the field on these problems. Finally, we also quantify and discuss the computational requirements needed for performing RL research by enumerating all experiments performed on these environments.
READ LESS

Summary

Some of the most important publications in deep reinforcement learning over the last few years have been fueled by access to massive amounts of computation through large scale distributed systems. The success of these approaches in achieving human-expert level performance on several complex video-game environments has motivated further exploration into...

READ MORE

Fast training of deep neural networks robust to adversarial perturbations

Published in:
2020 IEEE High Performance Extreme Computing Conf., HPEC, 22-24 September 2020.

Summary

Deep neural networks are capable of training fast and generalizing well within many domains. Despite their promising performance, deep networks have shown sensitivities to perturbations of their inputs (e.g., adversarial examples) and their learned feature representations are often difficult to interpret, raising concerns about their true capability and trustworthiness. Recent work in adversarial training, a form of robust optimization in which the model is optimized against adversarial examples, demonstrates the ability to improve performance sensitivities to perturbations and yield feature representations that are more interpretable. Adversarial training, however, comes with an increased computational cost over that of standard (i.e., nonrobust) training, rendering it impractical for use in largescale problems. Recent work suggests that a fast approximation to adversarial training shows promise for reducing training time and maintaining robustness in the presence of perturbations bounded by the infinity norm. In this work, we demonstrate that this approach extends to the Euclidean norm and preserves the human-aligned feature representations that are common for robust models. Additionally, we show that using a distributed training scheme can further reduce the time to train robust deep networks. Fast adversarial training is a promising approach that will provide increased security and explainability in machine learning applications for which robust optimization was previously thought to be impractical.
READ LESS

Summary

Deep neural networks are capable of training fast and generalizing well within many domains. Despite their promising performance, deep networks have shown sensitivities to perturbations of their inputs (e.g., adversarial examples) and their learned feature representations are often difficult to interpret, raising concerns about their true capability and trustworthiness. Recent...

READ MORE

Deep implicit coordination graphs for multi-agent reinforcement learning [e-print]

Summary

Multi-agent reinforcement learning (MARL) requires coordination to efficiently solve certain tasks. Fully centralized control is often infeasible in such domains due to the size of joint action spaces. Coordination graph based formalization allows reasoning about the joint action based on the structure of interactions. However, they often require domain expertise in their design. This paper introduces the deep implicit coordination graph (DICG) architecture for such scenarios. DICG consists of a module for inferring the dynamic coordination graph structure which is then used by a graph neural network based module to learn to implicitly reason about the joint actions or values. DICG allows learning the tradeoff between full centralization and decentralization via standard actor-critic methods to significantly improve coordination for domains with large number of agents. We apply DICG to both centralized-training-centralized-execution and centralized-training-decentralized-execution regimes. We demonstrate that DICG solves the relative overgeneralization pathology in predatory-prey tasks as well as outperforms various MARL baselines on the challenging StarCraft II Multi-agent Challenge (SMAC) and traffic junction environments.
READ LESS

Summary

Multi-agent reinforcement learning (MARL) requires coordination to efficiently solve certain tasks. Fully centralized control is often infeasible in such domains due to the size of joint action spaces. Coordination graph based formalization allows reasoning about the joint action based on the structure of interactions. However, they often require domain expertise...

READ MORE

Toward an autonomous aerial survey and planning system for humanitarian aid and disaster response

Summary

In this paper we propose an integrated system concept for autonomously surveying and planning emergency response for areas impacted by natural disasters. Referred to as AASAPS-HADR, this system is composed of a network of ground stations and autonomous aerial vehicles interconnected by an ad hoc emergency communication network. The system objectives are three-fold: to provide situational awareness of the evolving disaster event, to generate dispatch and routing plans for emergency vehicles, and to provide continuous communication networks which augment pre-existing communication infrastructure that may have been damaged or destroyed. Lacking development in previous literature, we give particular emphasis to the situational awareness objective of disaster response by proposing an autonomous aerial survey that is tasked with assessing damage to existing road networks, detecting and locating human victims, and providing a cursory assessment of casualty types that can be used to inform medical response priorities. In this paper we provide a high-level system design concept, identify existing AI perception and planning algorithms that most closely suit our purposes as well as technology gaps within those algorithms, and provide initial experimental results for non-contact health monitoring using real-time pose recognition algorithms running on a Nvidia Jetson TX2 mounted on board a quadrotor UAV. Finally we provide technology development recommendations for future phases of the AASAPS-HADR system.
READ LESS

Summary

In this paper we propose an integrated system concept for autonomously surveying and planning emergency response for areas impacted by natural disasters. Referred to as AASAPS-HADR, this system is composed of a network of ground stations and autonomous aerial vehicles interconnected by an ad hoc emergency communication network. The system...

READ MORE

Safe predictors for enforcing input-output specifications [e-print]

Summary

We present an approach for designing correct-by-construction neural networks (and other machine learning models) that are guaranteed to be consistent with a collection of input-output specifications before, during, and after algorithm training. Our method involves designing a constrained predictor for each set of compatible constraints, and combining them safely via a convex combination of their predictions. We demonstrate our approach on synthetic datasets and an aircraft collision avoidance problem.
READ LESS

Summary

We present an approach for designing correct-by-construction neural networks (and other machine learning models) that are guaranteed to be consistent with a collection of input-output specifications before, during, and after algorithm training. Our method involves designing a constrained predictor for each set of compatible constraints, and combining them safely via...

READ MORE

Graph matching via multi-scale heat diffusion

Author:
Published in:
IEEE Intl. Conf. on Big Data, 9-12 December 2019.

Summary

We propose a novel graph matching algorithm that uses ideas from graph signal processing to match vertices of graphs using alternative graph representations. Specifically, we consider a multi-scale heat diffusion on the graphs to create multiple weighted graph representations that incorporate both direct adjacencies as well as local structures induced from the heat diffusion. Then a multi-objective optimization method is used to match vertices across all pairs of graph representations simultaneously. We show that our proposed algorithm performs significantly better than the algorithm that only uses the adjacency matrices, especially when the number of known latent alignments between vertices (seeds) is small. We test the algorithm on a set of graphs and show that at the low seed level, the proposed algorithm performs at least 15–35% better than the traditional graph matching algorithm.
READ LESS

Summary

We propose a novel graph matching algorithm that uses ideas from graph signal processing to match vertices of graphs using alternative graph representations. Specifically, we consider a multi-scale heat diffusion on the graphs to create multiple weighted graph representations that incorporate both direct adjacencies as well as local structures induced...

READ MORE

Identification and detection of human trafficking using language models

Author:
Published in:
European Intelligence and Security Informatics Conf., EISIC, 26-27 November 2019.

Summary

In this paper, we present a novel language model-based method for detecting both human trafficking ads and trafficking indicators. The proposed system leverages language models to learn language structures in adult service ads, automatically select a list of keyword features, and train a machine learning model to detect human trafficking ads. The method is interpretable and adaptable to changing keywords used by traffickers. We apply this method to the Trafficking-10k dataset and show that it achieves better results than the previous models that leverage both ad text and images for detection. Furthermore, we demonstrate that our system can be successfully applied to detect suspected human trafficking organizations and rank these organizations based on their risk scores. This method provides a powerful new capability for law enforcement to rapidly identify ads and organizations that are suspected of human trafficking and allow more proactive policing using data.
READ LESS

Summary

In this paper, we present a novel language model-based method for detecting both human trafficking ads and trafficking indicators. The proposed system leverages language models to learn language structures in adult service ads, automatically select a list of keyword features, and train a machine learning model to detect human trafficking...

READ MORE