A survey of safety and trustworthiness of large language models through the lens of verification and validation

X Huang, W Ruan, W Huang, G Jin, Y Dong… - Artificial Intelligence …, 2024 - Springer
Large language models (LLMs) have exploded a new heatwave of AI for their ability to
engage end-users in human-level conversations with detailed and articulate answers across …

Safe reinforcement learning for power system control: A review

P Yu, Z Wang, H Zhang, Y Song - arXiv preprint arXiv:2407.00681, 2024 - arxiv.org
The large-scale integration of intermittent renewable energy resources introduces increased
uncertainty and volatility to the supply side of power systems, thereby complicating system …

[HTML][HTML] Machine learning meets advanced robotic manipulation

S Nahavandi, R Alizadehsani, D Nahavandi, CP Lim… - Information …, 2024 - Elsevier
Automated industries lead to high quality production, lower manufacturing cost and better
utilization of human resources. Robotic manipulator arms have major role in the automation …

Safe Multiagent Learning With Soft Constrained Policy Optimization in Real Robot Control

S Gu, D Huang, M Wen, G Chen… - IEEE Transactions on …, 2024 - ieeexplore.ieee.org
Due to a lack of safety considerations, a wide range of multiagent reinforcement learning
(MARL) applications are limited in real-world environments. Thus, ensuring MARL safety is …

Balance reward and safety optimization for safe reinforcement learning: A perspective of gradient manipulation

S Gu, B Sel, Y Ding, L Wang, Q Lin, M Jin… - Proceedings of the AAAI …, 2024 - ojs.aaai.org
Ensuring the safety of Reinforcement Learning (RL) is crucial for its deployment in real-world
applications. Nevertheless, managing the trade-off between reward and safety during …

Mutual Enhancement of Large Language and Reinforcement Learning Models through Bi-Directional Feedback Mechanisms: A Case Study

S Gu - arXiv preprint arXiv:2401.06603, 2024 - arxiv.org
Large Language Models (LLMs) have demonstrated remarkable capabilities for
reinforcement learning (RL) models, such as planning and reasoning capabilities. However …

ROSCOM: Robust Safe Reinforcement Learning on Stochastic Constraint Manifolds

S Gu, P Liu, A Kshirsagar, G Chen… - IEEE Transactions on …, 2024 - ieeexplore.ieee.org
Reinforcement Learning (RL) has demonstrated remarkable success across various
domains. Nonetheless, a significant challenge in RL is to ensure safety, particularly when …

Safe Multi-Agent Reinforcement Learning with Bilevel Optimization in Autonomous Driving

Z Zheng, S Gu - arXiv preprint arXiv:2405.18209, 2024 - arxiv.org
Ensuring safety in MARL, particularly when deploying it in real-world applications such as
autonomous driving, emerges as a critical challenge. To address this challenge, traditional …

SCPO: Safe Reinforcement Learning with Safety Critic Policy Optimization

J Mhamed, S Gu - arXiv preprint arXiv:2311.00880, 2023 - arxiv.org
Incorporating safety is an essential prerequisite for broadening the practical applications of
reinforcement learning in real-world scenarios. To tackle this challenge, Constrained Markov …

Safe and Balanced: A Framework for Constrained Multi-Objective Reinforcement Learning

S Gu, B Sel, Y Ding, L Wang, Q Lin, A Knoll… - arXiv preprint arXiv …, 2024 - arxiv.org
In numerous reinforcement learning (RL) problems involving safety-critical systems, a key
challenge lies in balancing multiple objectives while simultaneously meeting all stringent …