1 code implementation • 28 Feb 2024 • Hongchao Zhang, Luyao Niu, Andrew Clark, Radha Poovendran
Control barrier function (CBF)-based approaches have been proposed to guarantee the safety of robotic systems.
1 code implementation • 19 Feb 2024 • Fengqing Jiang, Zhangchen Xu, Luyao Niu, Zhen Xiang, Bhaskar Ramasubramanian, Bo Li, Radha Poovendran
In this paper, we propose a novel ASCII art-based jailbreak attack and introduce a comprehensive benchmark Vision-in-Text Challenge (ViTC) to evaluate the capabilities of LLMs in recognizing prompts that cannot be solely interpreted by semantics.
1 code implementation • 14 Feb 2024 • Zhangchen Xu, Fengqing Jiang, Luyao Niu, Jinyuan Jia, Bill Yuchen Lin, Radha Poovendran
Our results show that SafeDecoding significantly reduces the attack success rate and harmfulness of jailbreak attacks without compromising the helpfulness of responses to benign user queries.
no code implementations • 12 Feb 2024 • Dinuka Sahabandu, Xiaojun Xu, Arezoo Rajabi, Luyao Niu, Bhaskar Ramasubramanian, Bo Li, Radha Poovendran
We propose and analyze an adaptive adversary that can retrain a Trojaned DNN and is also aware of SOTA output-based Trojaned model detectors.
no code implementations • 2 Feb 2024 • Arezoo Rajabi, Reeya Pimple, Aiswarya Janardhanan, Surudhi Asokraj, Bhaskar Ramasubramanian, Radha Poovendran
However, the suitability of TL as a solution to reduce vulnerability of overfitted DNNs to privacy attacks is unexplored.
1 code implementation • 20 Jan 2024 • Zhen Xiang, Fengqing Jiang, Zidi Xiong, Bhaskar Ramasubramanian, Radha Poovendran, Bo Li
Moreover, we show that LLMs endowed with stronger reasoning capabilities exhibit higher susceptibility to BadChain, exemplified by a high average attack success rate of 97. 0% across the six benchmark tasks on GPT-4.
no code implementations • 10 Jan 2024 • Zhangchen Xu, Fengqing Jiang, Luyao Niu, Jinyuan Jia, Radha Poovendran
Our results show that the global model learned with Brave in the presence of adversaries achieves comparable classification accuracy to a global model trained in the absence of any adversary.
no code implementations • 7 Nov 2023 • Fengqing Jiang, Zhangchen Xu, Luyao Niu, Boxin Wang, Jinyuan Jia, Bo Li, Radha Poovendran
Successful exploits of the identified vulnerabilities result in the users receiving responses tailored to the intent of a threat initiator.
1 code implementation • 30 Aug 2023 • Arezoo Rajabi, Surudhi Asokraj, Fengqing Jiang, Luyao Niu, Bhaskar Ramasubramanian, Jim Ritcey, Radha Poovendran
An adversary carrying out a backdoor attack embeds a predefined perturbation called a trigger into a small subset of input samples and trains the DNN such that the presence of the trigger in the input results in an adversary-desired output class.
no code implementations • 4 Apr 2023 • Abdullah Al Maruf, Luyao Niu, Bhaskar Ramasubramanian, Andrew Clark, Radha Poovendran
We then propose a distributed MARL algorithm called the CVaR QD-Learning algorithm, and establish that value functions of individual agents reaches consensus.
no code implementations • 3 Dec 2022 • Arezoo Rajabi, Dinuka Sahabandu, Luyao Niu, Bhaskar Ramasubramanian, Radha Poovendran
Overfitted models have been shown to be susceptible to query-based attacks such as membership inference attacks (MIAs).
no code implementations • 13 Jul 2022 • Dinuka Sahabandu, Arezoo Rajabi, Luyao Niu, Bo Li, Bhaskar Ramasubramanian, Radha Poovendran
The results show that (i) with Submodular Trojan algorithm, the adversary needs to embed a Trojan trigger into a very small fraction of samples to achieve high accuracy on both Trojan and clean samples, and (ii) the MM Trojan algorithm yields a trained Trojan model that evades detection with probability 1.
no code implementations • 13 Apr 2022 • Dinuka Sahabandu, Sukarno Mertoguno, Radha Poovendran
Empirical evaluations show that using our byte-level features in ML-based ISA identification results in an 8% higher accuracy than the state-of-the-art features based on byte-histograms and byte pattern signatures.
no code implementations • 25 Mar 2022 • Arezoo Rajabi, Bhaskar Ramasubramanian, Radha Poovendran
We term this category of attacks multi-target backdoor attacks.
no code implementations • 18 Mar 2022 • Arezoo Rajabi, Bhaskar Ramasubramanian, Abdullah Al Maruf, Radha Poovendran
Through empirical evaluations, we highlight a privacy-utility tradeoff and demonstrate that the RL agent is able to learn behaviors that are aligned with that of a human user in the same environment in a privacy-preserving manner
1 code implementation • 19 Feb 2022 • Baicen Xiao, Bhaskar Ramasubramanian, Radha Poovendran
We design two algorithms- Shaping Advice in Single-agent reinforcement learning (SAS) and Shaping Advice in Multi-agent reinforcement learning (SAM).
Multi-agent Reinforcement Learning reinforcement-learning +1
1 code implementation • 12 Jan 2022 • Baicen Xiao, Bhaskar Ramasubramanian, Radha Poovendran
In this paper, we introduce Agent-Temporal Attention for Reward Redistribution in Episodic Multi-Agent Reinforcement Learning (AREL) to address these two challenges.
Multi-agent Reinforcement Learning reinforcement-learning +2
no code implementations • 3 Aug 2021 • Luyao Niu, Dinuka Sahabandu, Andrew Clark, Radha Poovendran
In this paper, we study the controlled islanding problem of a power system under disturbances introduced by a malicious adversary.
1 code implementation • 29 Mar 2021 • Baicen Xiao, Bhaskar Ramasubramanian, Radha Poovendran
We observe that using SAM results in agents learning policies to complete tasks faster, and obtain higher rewards than: i) using sparse rewards alone; ii) a state-of-the-art reward redistribution method.
Multi-agent Reinforcement Learning reinforcement-learning +1
no code implementations • 29 Mar 2021 • Bhaskar Ramasubramanian, Luyao Niu, Andrew Clark, Radha Poovendran
In this paper, we consider a setting where an autonomous agent has to learn behaviors in an unknown environment.
no code implementations • 20 Sep 2020 • Bhaskar Ramasubramanian, Baicen Xiao, Linda Bushnell, Radha Poovendran
We propose an iterative approach to the synthesis of the controller by solving a modified discrete-time Riccati equation.
1 code implementation • 24 Jul 2020 • Shana Moothedath, Dinuka Sahabandu, Joey Allen, Linda Bushnell, Wenke Lee, Radha Poovendran
Our game model has imperfect information as the players do not have information about the actions of the opponent.
Computer Science and Game Theory Cryptography and Security
no code implementations • 19 Jan 2020 • Baicen Xiao, Qifan Lu, Bhaskar Ramasubramanian, Andrew Clark, Linda Bushnell, Radha Poovendran
The output of the feedback neural network is converted to a shaping reward that is augmented to the reward provided by the environment.
no code implementations • 28 Jul 2019 • Hossein Hosseini, Sreeram Kannan, Radha Poovendran
In this paper, we first develop a classifier-based adaptation of the statistical test method and show that it improves the detection performance.
no code implementations • 20 Jul 2019 • Baicen Xiao, Bhaskar Ramasubramanian, Andrew Clark, Hannaneh Hajishirzi, Linda Bushnell, Radha Poovendran
This paper augments the reward received by a reinforcement learning agent with potential functions in order to help the agent learn (possibly stochastic) optimal policies.
no code implementations • 1 May 2019 • Hossein Hosseini, Sreeram Kannan, Radha Poovendran
Deep neural networks are vulnerable against adversarial examples.
no code implementations • 21 Mar 2018 • Hossein Hosseini, Baicen Xiao, Mayoore Jaiswal, Radha Poovendran
In order to conduct large scale experiments, we propose using the model accuracy on images with reversed brightness as a metric to evaluate the shape bias property.
1 code implementation • 16 Mar 2018 • Hossein Hosseini, Radha Poovendran
This property is used by several defense methods to counter adversarial examples by applying denoising filters or training the model to be robust to small perturbations.
no code implementations • 14 Aug 2017 • Hossein Hosseini, Baicen Xiao, Andrew Clark, Radha Poovendran
At the end, we propose introducing randomness to video analysis algorithms as a countermeasure to our attacks.
no code implementations • 16 Apr 2017 • Hossein Hosseini, Baicen Xiao, Radha Poovendran
For example, an adversary can bypass an image filtering system by adding noise to inappropriate images.
no code implementations • 26 Mar 2017 • Hossein Hosseini, Baicen Xiao, Radha Poovendran
For this, we select an image, which is different from the video content, and insert it, periodically and at a very low rate, into the video.
no code implementations • 20 Mar 2017 • Hossein Hosseini, Baicen Xiao, Mayoore Jaiswal, Radha Poovendran
To this end, we evaluate CNNs on negative images, since they share the same structure and semantics as regular images and humans can classify them correctly.
no code implementations • 13 Mar 2017 • Hossein Hosseini, Yize Chen, Sreeram Kannan, Baosen Zhang, Radha Poovendran
Advances in Machine Learning (ML) have led to its adoption as an integral component in many applications, including banking, medical diagnosis, and driverless cars.
no code implementations • 27 Feb 2017 • Hossein Hosseini, Sreeram Kannan, Baosen Zhang, Radha Poovendran
In this paper, we propose an attack on the Perspective toxic detection system based on the adversarial examples.
no code implementations • 27 Aug 2016 • Hossein Hosseini, Sreeram Kannan, Baosen Zhang, Radha Poovendran
We consider the setting where a collection of time series, modeled as random processes, evolve in a causal manner, and one is interested in learning the graph governing the relationships of these processes.
no code implementations • 28 Feb 2015 • Weiyao Lin, Ming-Ting Sun, Radha Poovendran, Zhengyou Zhang
This paper presents a novel approach for automatic recognition of group activities for video surveillance applications.
no code implementations • 28 Feb 2015 • Weiyao Lin, Ming-Ting Sun, Radha Poovendran, Zhengyou Zhang
This paper presents a novel approach for automatic recognition of human activities for video surveillance applications.