Perimeter Defense Problem

The perimeter defense problem focuses on optimizing resource allocation to protect a defined area from intrusions, whether these are network traffic congestion or physical threats. Current research explores diverse approaches, including decentralized multi-agent systems employing reinforcement learning and spike-based neural networks, as well as optimization techniques that prioritize threats based on factors like speed and location. These advancements aim to improve efficiency and robustness in managing perimeter security, with applications ranging from urban traffic management to cybersecurity and physical defense systems.

Papers

September 1, 2024

Generalized Multi-hop Traffic Pressure for Heterogeneous Traffic Perimeter Control
Xiaocan Li, Xiaoyu Wang, Ilia Smirnov, Scott Sanner, Baher Abdulhai
Multi Hop Mixed Traffic Heterogeneous Traffic Perimeter Defense Perimeter Defense Problem

July 19, 2023

Priority-based DREAM Approach for Highly Manoeuvring Intruders in A Perimeter Defense Problem
Shridhar Velhal, Suresh Sundaram, Narasimhan Sundararajan
Resource Allocation Priority Based Perimeter Defense Problem Arriving Intruder

May 26, 2023

A Decentralized Spike-based Learning Framework for Sequential Capture in Discrete Perimeter Defense Problem
Mohammed Thousif, Shridhar Velhal, Suresh Sundaram, Shirin Dora
Decentralized Learning Single Neural Network Multiple Defender Capture System Synaptic Change Perimeter Defense Problem

September 13, 2022

Data efficient reinforcement learning and adaptive optimal perimeter control of network traffic dynamics
C. Chen, Y. P. Huang, W. H. K. Lam, T. L. Pan, S. C. Hsu, A. Sumalee, R. X. Zhong
Raw Data Efficient Reinforcement Learning Network Traffic Traffic Dynamic Integral Reinforcement Learning Perimeter Defense Problem

November 26, 2021

The Geometry of Adversarial Training in Binary Classification
Leon Bungert, Nicolás García Trillos, Ryan Murray
Adversarial Attack Adversarial Training Binary Classification Geometric Analysis Convex Relaxation Learning Problem Nonlocal Perimeter Perimeter Defense Problem

Perimeter Defense Problem

Papers

Generalized Multi-hop Traffic Pressure for Heterogeneous Traffic Perimeter Control

Priority-based DREAM Approach for Highly Manoeuvring Intruders in A Perimeter Defense Problem

A Decentralized Spike-based Learning Framework for Sequential Capture in Discrete Perimeter Defense Problem

Data efficient reinforcement learning and adaptive optimal perimeter control of network traffic dynamics

The Geometry of Adversarial Training in Binary Classification