Substantial efforts have been made in developing various Decision Modeling
formalisms, both from industry and academia. A challenging problem is that of
expressing decision knowledge in the context of incomplete knowledge. In such
contexts, decisions depend on what is known or not known. We argue that none of
the existing formalisms for modeling decisions are capable of correctly
capturing the epistemic nature of such decisions, inevitably causing issues in
situations of uncertainty. This paper presents a new language for modeling
decisions with incomplete knowledge. It combines three principles:
stratification, autoepistemic logic, and definitions. A knowledge base in this
language is a hierarchy of epistemic theories, where each component theory may
epistemically reason on the knowledge in lower theories, and decisions are made
using definitions with epistemic conditions.

本文提出了一种用于对不完全知识建模决策的新语言，它结合了分层、自知逻辑和定义三个原则，其中知识库是一个认识论理论的层级结构，每个组成部分理论都可以对较低理论中的知识进行认识论推理，并使用具有认识条件的定义来进行决策。

一种建模完整知识上下文中决策的认知逻辑

An epistemic logic for modeling decisions in the context of incomplete  knowledge

Incomplete knowledge of the environment leads an agent to make decisions
under uncertainty. One of the major dilemmas in Reinforcement Learning (RL)
where an autonomous agent has to balance two contrasting needs in making its
decisions is: exploiting the current knowledge of the environment to maximize
the cumulative reward as well as exploring actions that allow improving the
knowledge of the environment, hopefully leading to higher reward values
(exploration-exploitation trade-off). Concurrently, another relevant issue
regards the full observability of the states, which may not be assumed in all
applications. Such as when only 2D images are considered as input in a RL
approach used for finding the optimal action within a 3D simulation
environment. In this work, we address these issues by deploying and testing
several techniques to balance exploration and exploitation trade-off on
partially observable systems for predicting steering wheels in autonomous
driving scenario. More precisely, the final aim is to investigate the effects
of using both stochastic and deterministic multi-armed bandit strategies
coupled with a Deep Recurrent Q-Network. Additionally, we adapted and evaluated
the impact of an innovative method to improve the learning phase of the
underlying Convolutional Recurrent Neural Network. We aim to show that adaptive
stochastic methods for exploration better approximate the trade-off between
exploration and exploitation as, in general, Softmax and Max-Boltzmann
strategies are able to outperform epsilon-greedy techniques.

使用深度递归 Q - 网络和多臂赌博策略，研究自主驾驶情景中平衡探索和利用的方法，以及部分可观测系统中预测方向盘的影响。

多臂赌博机策略对深度循环强化学习的影响

Impact of multi-armed bandit strategies on deep recurrent reinforcement  learning

In real-world applications, the ability to reason about incomplete knowledge,
sensing, temporal notions, and numeric constraints is vital. While several AI
planners are capable of dealing with some of these requirements, they are
mostly limited to problems with specific types of constraints. This paper
presents a new planning approach that combines contingent plan construction
within a temporal planning framework, offering solutions that consider numeric
constraints and incomplete knowledge. We propose a small extension to the
Planning Domain Definition Language (PDDL) to model (i) incomplete, (ii)
knowledge sensing actions that operate over unknown propositions, and (iii)
possible outcomes from non-deterministic sensing effects. We also introduce a
new set of planning domains to evaluate our solver, which has shown good
performance on a variety of problems.

本文提出了一种结合 PDDL 的新型规划方法，考虑了不完整知识、感知、时间和数字限制条件，并在新的一组规划领域上评估了我们的求解器。