[HTML][HTML] Decision-making under uncertainty: beyond probabilities: Challenges and perspectives
This position paper reflects on the state-of-the-art in decision-making under uncertainty. A
classical assumption is that probabilities can sufficiently capture all uncertainty in a system …
classical assumption is that probabilities can sufficiently capture all uncertainty in a system …
Confidence-aware reinforcement learning for self-driving cars
Reinforcement learning (RL) can be used to design smart driving policies in complex
situations where traditional methods cannot. However, they are frequently black-box in …
situations where traditional methods cannot. However, they are frequently black-box in …
Neural simplex architecture
Abstract We present the Neural Simplex Architecture (NSA), a new approach to runtime
assurance that provides safety guarantees for neural controllers (obtained eg using …
assurance that provides safety guarantees for neural controllers (obtained eg using …
Safe policy improvement for POMDPs via finite-state controllers
We study safe policy improvement (SPI) for partially observable Markov decision processes
(POMDPs). SPI is an offline reinforcement learning (RL) problem that assumes access to (1) …
(POMDPs). SPI is an offline reinforcement learning (RL) problem that assumes access to (1) …
[PDF][PDF] Alwayssafe: Reinforcement learning without safety constraint violations during training
Deploying reinforcement learning (RL) involves major concerns around safety. Engineering
a reward signal that allows the agent to maximize its performance while remaining safe is …
a reward signal that allows the agent to maximize its performance while remaining safe is …
Scalable safe policy improvement via Monte Carlo tree search
Algorithms for safely improving policies are important to deploy reinforcement learning
approaches in real-world scenarios. In this work, we propose an algorithm, called MCTS …
approaches in real-world scenarios. In this work, we propose an algorithm, called MCTS …
Partially Observable Monte Carlo Planning with state variable constraints for mobile robot navigation
Autonomous mobile robots employed in industrial applications often operate in complex and
uncertain environments. In this paper we propose an approach based on an extension of …
uncertain environments. In this paper we propose an approach based on an extension of …
[HTML][HTML] Efficient and scalable reinforcement learning for large-scale network control
The primary challenge in the development of large-scale artificial intelligence (AI) systems
lies in achieving scalable decision-making—extending the AI models while maintaining …
lies in achieving scalable decision-making—extending the AI models while maintaining …
Safe policy improvement with soft baseline bootstrapping
Abstract Batch Reinforcement Learning (Batch RL) consists in training a policy using
trajectories collected with another policy, called the behavioural policy. Safe policy …
trajectories collected with another policy, called the behavioural policy. Safe policy …
Identify, estimate and bound the uncertainty of reinforcement learning for autonomous driving
Deep reinforcement learning (DRL) has emerged as a promising approach for developing
more intelligent autonomous vehicles (AVs). A typical DRL application on AVs is to train a …
more intelligent autonomous vehicles (AVs). A typical DRL application on AVs is to train a …