no code implementations • 10 May 2024 • David "davidad" Dalrymple, Joar Skalse, Yoshua Bengio, Stuart Russell, Max Tegmark, Sanjit Seshia, Steve Omohundro, Christian Szegedy, Ben Goldhaber, Nora Ammann, Alessandro Abate, Joe Halpern, Clark Barrett, Ding Zhao, Tan Zhi-Xuan, Jeannette Wing, Joshua Tenenbaum
Ensuring that AI systems reliably and robustly avoid harmful or dangerous behaviours is a crucial challenge, especially for AI systems with a high degree of autonomy and general intelligence, or systems used in safety-critical contexts.
no code implementations • 29 Apr 2024 • Scott Viteri, Max Lamparth, Peter Chatain, Clark Barrett
We formalize the idea that the truthfulness of a sender to a receiver LM is the degree to which the sender helps the receiver predict their future observations.
1 code implementation • 25 Jan 2024 • Haoze Wu, Omri Isac, Aleksandar Zeljić, Teruhiro Tagomori, Matthew Daggitt, Wen Kokke, Idan Refaeli, Guy Amir, Kyle Julian, Shahaf Bassan, Pei Huang, Ori Lahav, Min Wu, Min Zhang, Ekaterina Komendantskaya, Guy Katz, Clark Barrett
This paper serves as a comprehensive system description of version 2. 0 of the Marabou framework for formal analysis of neural networks.
1 code implementation • 20 Dec 2023 • Pei Huang, Haoze Wu, Yuting Yang, Ieva Daukantas, Min Wu, Yedi Zhang, Clark Barrett
Quantization replaces floating point arithmetic with integer arithmetic in deep neural network models, providing more efficient on-device inference with less power and memory.
1 code implementation • 12 Dec 2023 • Lianmin Zheng, Liangsheng Yin, Zhiqiang Xie, Jeff Huang, Chuyue Sun, Cody Hao Yu, Shiyi Cao, Christos Kozyrakis, Ion Stoica, Joseph E. Gonzalez, Clark Barrett, Ying Sheng
SGLang is designed for the efficient programming of LLMs and incorporates primitives for common LLM programming patterns.
1 code implementation • 26 Oct 2023 • Chuyue Sun, Ying Sheng, Oded Padon, Clark Barrett
The use of large language models for code generation is a rapidly growing trend in software development.
1 code implementation • 7 Oct 2023 • Haoze Wu, Clark Barrett, Nina Narodytska
The demonstrated code-understanding capability of LLMs raises the question of whether they can be used for automated program verification, a task that demands high-level abstract reasoning about program properties that is challenging for verification tools.
no code implementations • 28 Aug 2023 • Clark Barrett, Brad Boyd, Elie Burzstein, Nicholas Carlini, Brad Chen, Jihye Choi, Amrita Roy Chowdhury, Mihai Christodorescu, Anupam Datta, Soheil Feizi, Kathleen Fisher, Tatsunori Hashimoto, Dan Hendrycks, Somesh Jha, Daniel Kang, Florian Kerschbaum, Eric Mitchell, John Mitchell, Zulfikar Ramzan, Khawaja Shams, Dawn Song, Ankur Taly, Diyi Yang
However, GenAI can be used just as well by attackers to generate new attacks and increase the velocity and efficacy of existing attacks.
1 code implementation • 24 Jun 2023 • Zhenyu Zhang, Ying Sheng, Tianyi Zhou, Tianlong Chen, Lianmin Zheng, Ruisi Cai, Zhao Song, Yuandong Tian, Christopher Ré, Clark Barrett, Zhangyang Wang, Beidi Chen
Based on these insights, we propose Heavy Hitter Oracle (H$_2$O), a KV cache eviction policy that dynamically retains a balance of recent and H$_2$ tokens.
1 code implementation • 3 Jun 2023 • Banghua Zhu, Ying Sheng, Lianmin Zheng, Clark Barrett, Michael I. Jordan, Jiantao Jiao
Theoretically, we provide an optimal algorithm for jointly optimizing both approaches to reduce the inference cost in both offline and online tabular settings.
1 code implementation • 18 May 2023 • Haoze Wu, Christopher Hahn, Florian Lonsing, Makai Mann, Raghuram Ramanujan, Clark Barrett
We present Self-Driven Strategy Learning ($\textit{sdsl}$), a lightweight online learning methodology for automated reasoning tasks that involve solving a set of related problems.
1 code implementation • 13 Mar 2023 • Ying Sheng, Lianmin Zheng, Binhang Yuan, Zhuohan Li, Max Ryabinin, Daniel Y. Fu, Zhiqiang Xie, Beidi Chen, Clark Barrett, Joseph E. Gonzalez, Percy Liang, Christopher Ré, Ion Stoica, Ce Zhang
As a result, when running OPT-175B on a single 16GB GPU, FlexGen achieves significantly higher throughput compared to state-of-the-art offloading systems, reaching a generation throughput of 1 token/s for the first time with an effective batch size of 144.
1 code implementation • 3 Mar 2023 • Dennis Wei, Haoze Wu, Min Wu, Pin-Yu Chen, Clark Barrett, Eitan Farchi
The softmax function is a ubiquitous component at the output of neural networks and increasingly in intermediate layers as well.
1 code implementation • 23 Oct 2022 • Elazar Cohen, Yizhak Yisrael Elboher, Clark Barrett, Guy Katz
Recent attempts have demonstrated that abstraction-refinement approaches could play a significant role in mitigating these limitations; but these approaches can often produce networks that are so abstract, that they become unsuitable for verification.
no code implementations • 16 Aug 2022 • Tom Zelazny, Haoze Wu, Clark Barrett, Guy Katz
A key component in many state-of-the-art verification schemes is computing lower and upper bounds on the values that neurons in the network can obtain for a specific input domain -- and the tighter these bounds, the more likely the verification is to succeed.
1 code implementation • 8 Jun 2022 • Haoze Wu, Teruhiro Tagomori, Alexander Robey, Fengjun Yang, Nikolai Matni, George Pappas, Hamed Hassani, Corina Pasareanu, Clark Barrett
We consider the problem of certifying the robustness of deep neural networks against real-world distribution shifts.
no code implementations • 1 Jun 2022 • Omri Isac, Clark Barrett, Min Zhang, Guy Katz
In this work, we present a novel mechanism for enhancing Simplex-based DNN verifiers with proof production capabilities: the generation of an easy-to-check witness of unsatisfiability, which attests to the absence of errors.
2 code implementations • 19 Mar 2022 • Haoze Wu, Aleksandar Zeljić, Guy Katz, Clark Barrett
Given a convex relaxation which over-approximates the non-convex activation functions, we encode the violations of activation functions as a cost function and optimize it with respect to the convex relaxation.
1 code implementation • 7 Mar 2022 • Haoze Wu, Clark Barrett, Mahmood Sharif, Nina Narodytska, Gagandeep Singh
Recently, Graph Neural Networks (GNNs) have been applied for scheduling jobs over clusters, achieving better performance than hand-crafted heuristics.
no code implementations • 6 Jan 2022 • Matan Ostrovsky, Clark Barrett, Guy Katz
Convolutional neural networks have gained vast popularity due to their excellent performance in the fields of computer vision, image processing, and others.
no code implementations • 2 Mar 2021 • Colin Paterson, Haoze Wu, John Grese, Radu Calinescu, Corina S. Pasareanu, Clark Barrett
We introduce DeepCert, a tool-supported method for verifying the robustness of deep neural network (DNN) image classifiers to contextually relevant perturbations such as blur, haze, and changes in image contrast.
1 code implementation • 5 Nov 2020 • Guy Amir, Haoze Wu, Clark Barrett, Guy Katz
One novelty of our technique is that it allows the verification of neural networks that include both binarized and non-binarized components.
no code implementations • 7 Oct 2020 • Christopher A. Strong, Haoze Wu, Aleksandar Zeljić, Kyle D. Julian, Guy Katz, Clark Barrett, Mykel J. Kochenderfer
However, individual "yes or no" questions cannot answer qualitative questions such as "what is the largest error within these bounds"; the answers to these lie in the domain of optimization.
no code implementations • 17 Apr 2020 • Haoze Wu, Alex Ozdemir, Aleksandar Zeljić, Ahmed Irfan, Kyle Julian, Divya Gopinath, Sadjad Fouladi, Guy Katz, Corina Pasareanu, Clark Barrett
Inspired by recent successes with parallel optimization techniques for solving Boolean satisfiability, we investigate a set of strategies and heuristics that aim to leverage parallel computing to improve the scalability of neural network verification.
1 code implementation • 6 Apr 2020 • Yuval Jacoby, Clark Barrett, Guy Katz
Deep neural networks are revolutionizing the way complex systems are developed.
1 code implementation • NeurIPS 2019 • Jiaxuan You, Haoze Wu, Clark Barrett, Raghuram Ramanujan, Jure Leskovec
The Boolean Satisfiability (SAT) problem is the canonical NP-complete problem and is fundamental to computer science, with a wide array of applications in planning, verification, and theorem proving.
no code implementations • 25 Oct 2019 • Sumathi Gokulanathan, Alexander Feldsher, Adi Malca, Clark Barrett, Guy Katz
Deep neural network (DNN) verification is an emerging field, with diverse verification engines quickly becoming available.
2 code implementations • 15 Mar 2019 • Changliu Liu, Tomer Arnon, Christopher Lazarus, Clark Barrett, Mykel J. Kochenderfer
Deep neural networks are widely used for nonlinear function approximation with applications ranging from computer vision to control.
no code implementations • 18 Jan 2018 • Lindsey Kuper, Guy Katz, Justin Gottschlich, Kyle Julian, Clark Barrett, Mykel Kochenderfer
The increasing use of deep neural networks for safety-critical applications, such as autonomous driving and flight control, raises concerns about their safety and reliability.
no code implementations • ICLR 2018 • Nicholas Carlini, Guy Katz, Clark Barrett, David L. Dill
We demonstrate how ground truths can serve to assess the effectiveness of attack techniques, by comparing the adversarial examples produced by those attacks to the ground truths; and also of defense techniques, by computing the distance to the ground truths before and after the defense is applied, and measuring the improvement.
no code implementations • 2 Oct 2017 • Divya Gopinath, Guy Katz, Corina S. Pasareanu, Clark Barrett
We propose a novel approach for automatically identifying safe regions of the input space, within which the network is robust against adversarial perturbations.
1 code implementation • 29 Sep 2017 • Nicholas Carlini, Guy Katz, Clark Barrett, David L. Dill
Using this approach, we demonstrate that one of the recent ICLR defense proposals, adversarial retraining, provably succeeds at increasing the distortion required to construct adversarial examples by a factor of 4. 2.
no code implementations • 8 Sep 2017 • Guy Katz, Clark Barrett, David L. Dill, Kyle Julian, Mykel J. Kochenderfer
Autonomous vehicles are highly complex systems, required to function reliably in a wide variety of situations.
8 code implementations • 3 Feb 2017 • Guy Katz, Clark Barrett, David Dill, Kyle Julian, Mykel Kochenderfer
Deep neural networks have emerged as a widely used and effective means for tackling complex, real-world problems.