大模型对齐与安全 - arXivDaily 专题

2009.11484 2020-09-25 cs.CR cs.AI cs.SE 57%

Pandora: A Cyber Range Environment for the Safe Testing and Deployment of Autonomous Cyber Attack Tools

Hetong Jiang, Taejun Choi, Ryan K. L. Ko

专题命中安全训练：safety(abstract)；分类 cs.AI

Comments 20 pages, 10 figures, to be published in SSCC 2020

2004.05273 2020-09-24 cs.RO cs.LG cs.MA cs.SY eess.SY math.OC 57%

Safe Multi-Agent Interaction through Robust Control Barrier Functions with Learned Uncertainties

Richard Cheng, Mohammad Javad Khojasteh, Aaron D. Ames, Joel W. Burdick

专题命中安全训练：safety(abstract)；分类 cs.LG

Journal ref 59th IEEE Conference on Decision and Control (CDC 2020)

2003.09398 2020-09-15 cs.LG cs.RO stat.ML 57%

Deep Constrained Q-learning

Gabriel Kalweit, Maria Huegle, Moritz Werling, Joschka Boedecker

专题命中安全训练：safety(abstract)；分类 cs.LG

2008.11811 2020-08-28 cs.LG math.OC stat.ML 57%

Constrained Markov Decision Processes via Backward Value Functions

Harsh Satija, Philip Amortila, Joelle Pineau

专题命中安全训练：safety(abstract)；分类 cs.LG

2006.05935 2020-06-11 cs.RO cs.LG 57%

Learning to Play Table Tennis From Scratch using Muscular Robots

Dieter Büchler, Simon Guist, Roberto Calandra, Vincent Berenz, Bernhard Schölkopf, Jan Peters

专题命中安全训练：safety(abstract)；分类 cs.LG

Comments 11 pages, 8 figures. Submitted to T-RO. For more information visit https://muscularTT.embodied.ml

2002.00434 2020-05-20 cs.AI 57%

Integrating Deep Reinforcement Learning with Model-based Path Planners for Automated Driving

Ekim Yurtsever, Linda Capito, Keith Redmill, Umit Ozguner

专题命中安全训练：safety(abstract)；分类 cs.AI

Comments 6 pages, 5 figures. Accepted for IEEE Intelligent Vehicles Symposium 2020

2001.03923 2020-04-07 cs.CV cs.LG eess.IV 57%

Bridging the gap between AI and Healthcare sides: towards developing clinically relevant AI-powered diagnosis systems

Changhee Han, Leonardo Rundo, Kohei Murao, Takafumi Nemoto, Hideki Nakayama

专题命中安全训练：safety(abstract)；分类 cs.LG

Comments 13 pages, 2 figure, accepted to AIAI 2020

1905.04194 2020-03-24 cs.LG stat.ML 57%

Formal Verification of Input-Output Mappings of Tree Ensembles

John Törnblom, Simin Nadjm-Tehrani

专题命中安全训练：safety(abstract)；分类 cs.LG

2002.12078 2020-02-28 cs.LG cs.RO cs.SY eess.SY stat.ML 57%

Training Adversarial Agents to Exploit Weaknesses in Deep Control Policies

Sampo Kuutti, Saber Fallah, Richard Bowden

专题命中安全训练：safety(abstract)；分类 cs.LG

Comments 2020 IEEE International Conference on Robotics and Automation (ICRA)

2002.11625 2020-02-27 cs.CY cs.CR cs.NI cs.SY eess.SY 57%

Death by AI: Where Assured Autonomy in Smart Cities Meets the End-to-End Argument

Gregory Falco

专题命中安全训练：safety(abstract)；分类 cs.CY

Comments PREPRINT, 6 Pages

2002.05615 2020-02-14 cs.AI 57%

Verifiable RNN-Based Policies for POMDPs Under Temporal Logic Constraints

Steven Carr, Nils Jansen, Ufuk Topcu

专题命中安全训练：safety(abstract)；分类 cs.AI

Comments 8 pages, 5 figures, 1 table

2002.04759 2020-02-13 cs.LG stat.ML 57%

Collaborative Inference for Efficient Remote Monitoring

Chi Zhang, Yong Sheng Soh, Ling Feng, Tianyi Zhou, Qianxiao Li

专题命中安全训练：safety(abstract)；分类 cs.LG

1806.07555 2020-01-28 cs.LG stat.ML 57%

Stagewise Safe Bayesian Optimization with Gaussian Processes

Yanan Sui, Vincent Zhuang, Joel W. Burdick, Yisong Yue

专题命中安全训练：safety(abstract)；分类 cs.LG

Comments International Conference on Machine Learning (ICML) 2018

2001.09113 2020-01-27 cs.RO cs.AI cs.SY eess.SY 57%

Perception as prediction using general value functions in autonomous driving applications

Daniel Graves, Kasra Rezaee, Sean Scheideman

专题命中安全训练：safety(abstract)；分类 cs.AI

Comments 8 pages, 6 figures, IROS 2019 conference

1912.12413 2020-01-01 stat.ML cs.LG 57%

Approval policies for modifications to Machine Learning-Based Software as a Medical Device: A study of bio-creep

Jean Feng, Scott Emerson, Noah Simon

专题命中安全训练：safety(abstract)；分类 cs.LG

1910.12639 2020-01-01 eess.SY cs.AI cs.MA cs.RO cs.SY 57%

MAMPS: Safe Multi-Agent Reinforcement Learning via Model Predictive Shielding

Wenbo Zhang, Osbert Bastani, Vijay Kumar

专题命中安全训练：safety(abstract)；分类 cs.AI

1906.08935 2019-12-20 cs.LG cs.CR stat.ML 57%

Deep Leakage from Gradients

Ligeng Zhu, Zhijian Liu, Song Han

专题命中安全训练：safety(abstract)；分类 cs.LG

1912.04066 2019-12-10 eess.SY cs.LG cs.SY math.OC 57%

Feasibility-Guided Learning for Robust Control in Constrained Optimal Control Problems

Wei Xiao, Calin A. Belta, Christos G. Cassandras

专题命中安全训练：safety(abstract)；分类 cs.LG

Comments 8 pages, submitted to L-CSS

1807.06096 2019-11-26 cs.AI 57%

Safe Reinforcement Learning via Probabilistic Shields

Nils Jansen, Bettina Könighofer, Sebastian Junges, Alexandru C. Serban, Roderick Bloem

专题命中安全训练：safety(abstract)；分类 cs.AI

1910.12156 2019-10-29 cs.LG stat.ML 57%

Convergent Policy Optimization for Safe Reinforcement Learning

Ming Yu, Zhuoran Yang, Mladen Kolar, Zhaoran Wang

专题命中安全训练：safety(abstract)；分类 cs.LG

1909.11228 2019-09-26 cs.LG stat.ML 57%

Avoidance Learning Using Observational Reinforcement Learning

David Venuto, Leonard Boussioux, Junhao Wang, Rola Dali, Jhelum Chakravorty, Yoshua Bengio, Doina Precup

专题命中安全训练：safety(abstract)；分类 cs.LG

1909.09540 2019-09-23 cs.LG stat.ML 57%

Reconnaissance and Planning algorithm for constrained MDP

Shin-ichi Maeda, Hayato Watahiki, Shintarou Okada, Masanori Koyama

专题命中安全训练：safety(abstract)；分类 cs.LG

1909.04886 2019-09-12 eess.SY cs.LG cs.SY stat.ML 57%

Towards Safe Machine Learning for CPS: Infer Uncertainty from Training Data

Xiaozhe Gu, Arvind Easwaran

专题命中安全训练：safety(abstract)；分类 cs.LG

Comments Publication rights licensed to ACM

Journal ref In Proceedings of the 10th ACM/IEEE International Conference on Cyber-Physical Systems (ICCPS), 2019. ACM, New York, NY, USA, pages 249-258

1908.10187 2019-08-28 cs.LG eess.SP stat.ML 57%

A Machine Learning Approach for Smartphone-based Sensing of Roads and Driving Style

M. Ricardo Carlos

专题命中安全训练：safety(abstract)；分类 cs.LG

Comments Doctoral Dissertation. Dissertation Advisors: Luis Carlos González Gurrola and Fernando Martínez

1908.08651 2019-08-26 cs.CY 57%

Trajectory-Based Urban Air Mobility (UAM) Operations Simulator (TUS)

Euclides C. Pinto Neto, Derick M. Baum, Jorge Rady de Almeida Junior, João Batista Camargo Junior, Paulo Sérgio Cugnasca

专题命中安全训练：safety(abstract)；分类 cs.CY

1902.08708 2019-06-17 stat.ML cs.LG 57%

Distributionally Robust Reinforcement Learning

Elena Smirnova, Elvis Dohmatob, Jérémie Mary

专题命中安全训练：safety(abstract)；分类 cs.LG

1810.00619 2019-06-17 cs.LG cs.PL stat.ML 57%

SmartChoices: Hybridizing Programming and Machine Learning

Victor Carbune, Thierry Coppey, Alexander Daryin, Thomas Deselaers, Nikhil Sarda, Jay Yagnik

专题命中安全训练：safety(abstract)；分类 cs.LG

Comments published at the Reinforcement Learning for Real Life (RL4RealLife) Workshop in the 36th International Conference on Machine Learning (ICML), Long Beach, California, USA, 2019

1906.02275 2019-06-07 cs.RO cs.LG stat.ML 57%

Continuous Control for Automated Lane Change Behavior Based on Deep Deterministic Policy Gradient Algorithm

Pin Wang, Hanhan Li, Ching-Yao Chan

专题命中安全训练：safety(abstract)；分类 cs.LG

Comments Published at the 30th IEEE Intelligent Vehicles Symposium (IV), 2019

1905.04223 2019-05-13 cs.LG cs.SE stat.ML 57%

Assuring the Machine Learning Lifecycle: Desiderata, Methods, and Challenges

Rob Ashmore, Radu Calinescu, Colin Paterson

专题命中安全训练：safety(abstract)；分类 cs.LG

1810.12380 2019-05-13 cs.CR cs.LG 57%

Conditionals in Homomorphic Encryption and Machine Learning Applications

Diego Chialva, Ann Dooms

专题命中安全训练：safety(abstract)；分类 cs.LG

Comments 14 pages, 1 figure, corrected typos, added introductory pedagogical section on polynomial approximation

1. 安全训练 3281 篇

Pandora: A Cyber Range Environment for the Safe Testing and Deployment of Autonomous Cyber Attack Tools

Safe Multi-Agent Interaction through Robust Control Barrier Functions with Learned Uncertainties

Deep Constrained Q-learning

Constrained Markov Decision Processes via Backward Value Functions

Learning to Play Table Tennis From Scratch using Muscular Robots

Integrating Deep Reinforcement Learning with Model-based Path Planners for Automated Driving

Bridging the gap between AI and Healthcare sides: towards developing clinically relevant AI-powered diagnosis systems

Formal Verification of Input-Output Mappings of Tree Ensembles

Training Adversarial Agents to Exploit Weaknesses in Deep Control Policies

Death by AI: Where Assured Autonomy in Smart Cities Meets the End-to-End Argument

Verifiable RNN-Based Policies for POMDPs Under Temporal Logic Constraints

Collaborative Inference for Efficient Remote Monitoring

Stagewise Safe Bayesian Optimization with Gaussian Processes

Perception as prediction using general value functions in autonomous driving applications

Approval policies for modifications to Machine Learning-Based Software as a Medical Device: A study of bio-creep

MAMPS: Safe Multi-Agent Reinforcement Learning via Model Predictive Shielding

Deep Leakage from Gradients

Feasibility-Guided Learning for Robust Control in Constrained Optimal Control Problems

Safe Reinforcement Learning via Probabilistic Shields

Convergent Policy Optimization for Safe Reinforcement Learning

Avoidance Learning Using Observational Reinforcement Learning

Reconnaissance and Planning algorithm for constrained MDP

Towards Safe Machine Learning for CPS: Infer Uncertainty from Training Data

A Machine Learning Approach for Smartphone-based Sensing of Roads and Driving Style

Trajectory-Based Urban Air Mobility (UAM) Operations Simulator (TUS)

Distributionally Robust Reinforcement Learning

SmartChoices: Hybridizing Programming and Machine Learning

Continuous Control for Automated Lane Change Behavior Based on Deep Deterministic Policy Gradient Algorithm

Assuring the Machine Learning Lifecycle: Desiderata, Methods, and Challenges

Conditionals in Homomorphic Encryption and Machine Learning Applications