同步wx订阅号(arXiv每日论文速递),支持后台回复'search 关键词'查询相关的最新论文。有些许帮助的话,麻烦关注一下哦(* ̄rǒ ̄)
cs.AI 方向,今日共计15篇
[cs.AI]:
【1】 Categorizing Wireheading in Partially Embedded Agents
标题:对部分嵌入代理中的线头进行分类
作者: Arushi Majha, Davide Zagami
备注:Accepted at the AI Safety Workshop in IJCAI 2019
链接:https://arxiv.org/abs/1906.09136
【2】 Hybrid Planning for Dynamic Multimodal Stochastic Shortest Paths
标题:动态多模态随机最短路径的混合规划
作者: Shushman Choudhury, Mykel J. Kochenderfer
备注:20 pages, 5 figures, 5 tables; Under Review
链接:https://arxiv.org/abs/1906.09094
【3】 Customer Segmentation of Wireless Trajectory Data
标题:无线轨迹数据的客户细分
作者: Matthew R Karlsen, Sotiris K. Moschoyiannis
备注:Technical Report, University of Surrey, UK, 2018
链接:https://arxiv.org/abs/1906.08874
【4】 Shaping Belief States with Generative Environment Models for RL
标题:用RL的生成环境模型塑造信念状态
作者: Karol Gregor, Aaron van den Oord
链接:https://arxiv.org/abs/1906.09237
【5】 Disentangled Skill Embeddings for Reinforcement Learning
标题:用于强化学习的解缠技能嵌入
作者: Janith C. Petangoda, Jordi Grau-Moya
链接:https://arxiv.org/abs/1906.09223
【6】 Mitigating Bias in Algorithmic Employment Screening: Evaluating Claims and Practices
标题:减轻算法就业筛选中的偏见:评估索赔和实践
作者: Manish Raghavan, Karen Levy
链接:https://arxiv.org/abs/1906.09208
【7】 Continual Reinforcement Learning with Diversity Exploration and Adversarial Self-Correction
标题:具有多样性探索和对抗性自校正的连续强化学习
作者: Fengda Zhu, Mingkui Tan
链接:https://arxiv.org/abs/1906.09205
【8】 Explainable Fact Checking with Probabilistic Answer Set Programming
标题:基于概率答案集编程的可解释事实检验
作者: Naser Ahmadi, Mohammed Saeed
链接:https://arxiv.org/abs/1906.09198
【9】 Near-optimal Reinforcement Learning using Bayesian Quantiles
标题:基于贝叶斯分位数的近似最优强化学习
作者: Aristide Tossou, Christos Dimitrakakis
备注:arXiv admin note: substantial text overlap with arXiv:1905.12425
链接:https://arxiv.org/abs/1906.09114
【10】 Learning Reward Functions by Integrating Human Demonstrations and Preferences
标题:通过集成人类演示和偏好来学习奖励函数
作者: Malayandi Palan, Dorsa Sadigh
备注:Presented at RSS 2019
链接:https://arxiv.org/abs/1906.08928
【11】 Variable Impedance Control in End-Effector Space: An Action Space for Reinforcement Learning in Contact-Rich Tasks
标题:末端效应器空间中的可变阻抗控制:富接触任务中用于强化学习的动作空间
作者: Roberto Martín-Martín, Animesh Garg
备注:IROS19
链接:https://arxiv.org/abs/1906.08880
【12】 Evolving Self-supervised Neural Networks: Autonomous Intelligence from Evolved Self-teaching
标题:进化的自我监督神经网络:来自进化自我教学的自主智能
作者: Nam Le
链接:https://arxiv.org/abs/1906.08865
【13】 Deep Neuroevolution of Recurrent and Discrete World Models
标题:递归和离散世界模型的深部神经进化
作者: Sebastian Risi, Kenneth O. Stanley
链接:https://arxiv.org/abs/1906.08857
【14】 A Deep Reinforcement Learning Approach for Global Routing
标题:一种全局路由的深度强化学习方法
作者: Haiguang Liao, Levent Burak Kara
链接:https://arxiv.org/abs/1906.08809
【15】 Finding Needles in a Moving Haystack: Prioritizing Alerts with Adversarial Reinforcement Learning
标题:在移动的草堆中寻找针:使用对抗性强化学习确定警报的优先顺序
作者: Liang Tong, Yevgeniy Vorobeychik
链接:https://arxiv.org/abs/1906.08805
翻译:腾讯翻译君