Skip to search formSkip to main contentSkip to account menu

Robust Multi-Agent Reinforcement Learning via Minimax Deep Deterministic Policy Gradient

This paper proposes a new algorithm, MiniMax Multi-agent Deep Deterministic Policy Gradient (M3DDPG) with the following contributions: a minimax extension of the popular multi-agent deep deterministic policy gradient algorithm (MADDPG), for robust policy learning; and a Multi-Agent Adversarial Learning (MAAL) to efficiently solve the proposed formulation.
View via Publisher (opens in a new tab)

Artificial Intelligence for Social Good: A Survey

This work quantitatively analyzes the distribution and trend of the AI4SG literature in terms of application domains and AI techniques used and proposes three conceptual methods to systematically group the existing literature and analyze the eight AI4 SG application domains in a unified framework.
View PDF on arXiv (opens in a new tab)

"A Game of Thrones": When Human Behavior Models Compete in Repeated Stackelberg Security Games

A new human behavior model, SHARP, which mitigates these three limitations as follows: (i) SHARP reasons based on success or failure of the adversary's past actions on exposed portions of the attack surface to model adversary adaptiveness; (ii)SHARP reasons about similarity between exposed and unexposed areas of the attacked surface, and incorporates a discounting parameter to mitigate adversary's lack of exposure to enough of theattack surface.
View Paper (opens in a new tab)

Evolutionary Population Curriculum for Scaling Multi-Agent Reinforcement Learning

EPC is introduced, a curriculum learning paradigm that scales up Multi-Agent Reinforcement Learning (MARL) by progressively increasing the population of training agents in a stage-wise manner and uses an evolutionary approach to fix an objective misalignment issue throughout the curriculum.
View PDF on arXiv (opens in a new tab)

When Security Games Go Green: Designing Defender Strategies to Prevent Poaching and Illegal Fishing

Green Security Games (GSGs) is introduced, a novel game model for green security domains with a generalized Stackelberg assumption; algorithms to plan effective sequential defender strategies are provided; a novel approach to learn adversary models that further improves defender performance is proposed.
View Paper (opens in a new tab)

Deploying PAWS: Field Optimization of the Protection Assistant for Wildlife Security

This paper reports on PAWS’s significant evolution from a proposed decision aid to a regularly deployed application, reporting on the lessons from the first tests in Africa in Spring 2014, through its continued evolution since then, to current regular use in Southeast Asia and plans for future worldwide deployment.
View via Publisher (opens in a new tab)

Stackelberg Security Games: Looking Beyond a Decade of Success

A broad survey of recent technical advances in Stackelberg Security Game and related literature is presented, and the future is highlighted by highlighting the new potential applications and open research problems in SSG.
View via Publisher (opens in a new tab)

Optimal patrol strategy for protecting moving targets with multiple mobile resources

This paper focuses on protecting mobile targets that lead to a continuous set of strategies for the players, motivated by several real-world domains including protecting ferries with escorts and protecting refugee supply lines.
View via Publisher (opens in a new tab)

What game are we playing? End-to-end learning in normal and extensive form games

A differentiable, end-to-end learning framework for addressing the relatively under-explored but equally important "inverse" setting, where the parameters of the underlying game are not known to all agents, but must be learned through observations.
View PDF on arXiv (opens in a new tab)

Deep Reinforcement Learning for Green Security Games with Real-Time Information

This work designs a novel deep reinforcement learning-based algorithm, DeDOL, to compute a patrolling strategy that adapts to the real-time information against a best-responding attacker, and is the first attempt to use Deep Q-Learning for security games.
View PDF on arXiv (opens in a new tab)
...
By clicking accept or continuing to use the site, you agree to the terms outlined in our Privacy Policy (opens in a new tab), Terms of Service (opens in a new tab), and Dataset License (opens in a new tab)