Gaussian Process Based Message Filtering for Robust Multi-Agent Cooperation in the Presence of Adversarial Communication

Rupert, Mitchell,; Blumenkamp, Jan; Prorok, Amanda

doi:10.48550/arxiv.2012.00508

Cited by 5 publications

(11 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We seek to minimize the probability of error (15) for the decision rule (12) by minimizing the false alarm and missed detection probabilities. Any sequence of 0's and 1's can occur for the detected trust vector t, each yielding a different error probability, so the error probability must be calculated for each possible vector t, along with each possible vector y.…”

Section: A Two Stage Approach Algorithmmentioning

confidence: 99%

See 1 more Smart Citation

Adaptive Malicious Robot Detection in Dynamic Topologies

Cavorsi

Jadhav

Saldaña

et al. 2022

2022 IEEE 61st Conference on Decision and Control (CDC)

View full text Add to dashboard Cite

We develop a resilient binary hypothesis testing framework for decision making in adversarial multi-robot crowdsensing tasks. This framework exploits stochastic trust observations between robots to arrive at tractable, resilient decision making at a centralized Fusion Center (FC) even when i) there exist malicious robots in the network and their number may be larger than the number of legitimate robots, and ii) the FC uses one-shot noisy measurements from all robots. We derive two algorithms to achieve this. The first is the Two Stage Approach (2SA) that estimates the legitimacy of robots based on received trust observations, and provably minimizes the probability of detection error in the worst-case malicious attack. Here, the proportion of malicious robots is known but arbitrary. For the case of an unknown proportion of malicious robots, we develop the Adversarial Generalized Likelihood Ratio Test (A-GLRT) that uses both the reported robot measurements and trust observations to estimate the trustworthiness of robots, their reporting strategy, and the correct hypothesis simultaneously. We exploit special problem structure to show that this approach remains computationally tractable despite several unknown problem parameters. We deploy both algorithms in a hardware experiment where a group of robots conducts crowdsensing of traffic conditions on a mock-up road network similar in spirit to Google Maps, subject to a Sybil attack. We extract the trust observations for each robot from actual communication signals which provide statistical information on the uniqueness of the sender. We show that even when the malicious robots are in the majority, the FC can reduce the probability of detection error to 30.5% and 29% for the 2SA and the A-GLRT respectively.

show abstract

Section: A Two Stage Approach Algorithmmentioning

confidence: 99%

“…Unfortunately, this computation scales exponentially with the number of robots, N . Furthermore, the true trust vector t and the probabilities of false alarm and missed detection of the malicious robots are unknown, i.e., P FA,M and P MD,M , therefore, they cannot be used in minimizing (15).…”

Section: A Two Stage Approach Algorithmmentioning

confidence: 99%

Adaptive Malicious Robot Detection in Dynamic Topologies

Cavorsi

Jadhav

Saldaña

et al. 2022

2022 IEEE 61st Conference on Decision and Control (CDC)

View full text Add to dashboard Cite

show abstract

“…As discussed in Sec. 1, recent works [2,21] are either focusing on attacking specific competitive setting with simple attack method without effective defense or defensing on specific communication method. Our work fill this gap in MACRL.…”

Section: Related Workmentioning

confidence: 99%

“…Besides, the defense methods had not been proposed. Mitchell et al [21] propose a defensive method in MACRL by replacing weights generated by Attention [36] model in MACRL method with weights generated by Gaussian process. However, such method is only restricted to attention-based MACRL method and neglects the scenarios where attacking strategies are learnable.…”

Section: Introductionmentioning

confidence: 99%

Mis-spoke or mis-lead: Achieving Robustness in Multi-Agent Communicative Reinforcement Learning

Xue

Qiu

et al. 2021

Preprint

View full text Add to dashboard Cite

Recent studies in multi-agent communicative reinforcement learning (MACRL) demonstrate that multi-agent coordination can be significantly improved when communication between agents is allowed. Meanwhile, advances in adversarial machine learning (ML) have shown that ML and reinforcement learning (RL) models are vulnerable to a variety of attacks that significantly degrade the performance of learned behaviours. However, despite the obvious and growing importance, the combination of adversarial ML and MACRL remains largely uninvestigated. In this paper, we make the first step towards conducting message attacks on MACRL methods. In our formulation, one agent in the cooperating group is taken over by an adversary and can send malicious messages to disrupt a deployed MACRLbased coordinated strategy during the deployment phase. We further our study by developing a defence method via message reconstruction. Finally, we address the resulting arms race, i.e., we consider the ability of the malicious agent to adapt to the changing and improving defensive communicative policies of the benign agents. Specifically, we model the adversarial MACRL problem as a two-player zero-sum game and then utilize Policy-Space Response Oracle to achieve communication robustness. Empirically, we demonstrate that MACRL methods are vulnerable to message attacks while our defence method the game-theoretic framework can effectively improve the robustness of MACRL. * Equal contribution.Preprint. Under review.

show abstract

“…ture, our method is also applicable to any scenario where multiple images need to be translated into one, such as human image generation [21] from multiple sources. In addition, we believe that our work can be explored in 3D reconstruction settings or in multi-agent communication settings where multiple drones share their image captures from different locations and construct a single coherent image (see [28] for a similar setting).…”

Section: Limitations and Failure Casesmentioning

confidence: 99%

Graph2Pix: A Graph-Based Image to Image Translation Framework

Gokay¹,

Simsar²,

Atici³

et al. 2021

Preprint

View full text Add to dashboard Cite

In this paper, we propose a graph-based image-to-image translation framework for generating images. We use rich data collected from the popular creativity platform Artbreeder 1 , where users interpolate multiple GAN-generated images to create artworks. This unique approach of creating new images leads to a tree-like structure where one can track historical data about the creation of a particular image. Inspired by this structure, we propose a novel graph-to-image translation model called Graph2Pix, which takes a graph and corresponding images as input and generates a single image as output. Our experiments show that Graph2Pix is able to outperform several image-toimage translation frameworks on benchmark metrics, including LPIPS (with a 25% improvement) and human perception studies (n = 60), where users preferred the images generated by our method 81.5% of the time. Our source code and dataset are publicly available at https: //github.com/catlab-team/graph2pix.

show abstract

Gaussian Process Based Message Filtering for Robust Multi-Agent Cooperation in the Presence of Adversarial Communication

Cited by 5 publications

References 22 publications

Adaptive Malicious Robot Detection in Dynamic Topologies

Adaptive Malicious Robot Detection in Dynamic Topologies

Mis-spoke or mis-lead: Achieving Robustness in Multi-Agent Communicative Reinforcement Learning

Graph2Pix: A Graph-Based Image to Image Translation Framework

Contact Info

Product

Resources

About