Human-Like Computing Machine Intelligence Workshop

类人计算机智能研讨会

• 批准号: EP/P006272/1
• 负责人: Stephen Muggleton
• 金额: $ 4.3万
• 依托单位: Imperial College London
• 依托单位国家: 英国
• 项目类别: Research Grant
• 财政年份: 2016
• 资助国家: 英国
• 起止时间: 2016 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=EP%2FP006272%2F1
• 关键词: Human; Like; Computing; Machine; Intelligence

Recent prominent advances in statistical AI, such as the success ofdriverless cars and Alpha Go's recent defeat ofthe 9-dan ranked South Korean Go player Lee Se-dol have considerablyincreased public awareness of the potential for computers to achievehuman-level competence in settings which were previously the exclusiverealm of human judgement and skill. However, in human termsstatistical AI techniques still act as skilled zombies, havingunconscious skills detached from the ability to express goals or intentions.The lack of ability to communicate in a co-operative fashion withhuman beings limits the long-term potential of these approachesin many commercially and socially valuable settings.The workshop will address issues related to interaction in which human co-operative and communicative skillscan be studied and modelled computationally. While many of these issues havebeen prominent for some time in symbolic AI, we are yetto see successful integration of statistical and symbolic AI approacheswhich achieves the broad range of phenomena present in human behaviour.This workshop aims to contribute to the objectives of the EPSRC's Human-LikeComputing priority area by crystalising the key issues which need tobe prioritised in this area.

统计人工智能领域最近取得的显著进展，如无人驾驶汽车的成功和阿尔法围棋最近击败了韩国九段围棋选手李世石，这无疑提高了公众对计算机在以前人类判断和技能的专属领域取代人类水平能力的潜力的认识。然而，在人类的术语统计人工智能技术仍然作为熟练的僵尸，具有与表达目标或意图的能力无关的无意识技能。缺乏与人类合作沟通的能力限制了这些方法在许多商业和社会价值环境中的长期潜力。研讨会将讨论与人类合作的互动相关的问题。操作和沟通技巧可以通过计算来研究和建模。虽然这些问题在符号人工智能中已经突出了一段时间，但我们还没有看到统计和符号人工智能方法的成功整合，从而实现人类行为中存在的广泛现象。本次研讨会旨在通过明确这一领域需要优先考虑的关键问题，为EPSRC的类人计算优先领域的目标做出贡献。

项目成果

Turning 30: New Ideas in Inductive Logic Programming

• DOI: 10.24963/ijcai.2020/673
• 发表时间: 2020-02
• 期刊: ArXiv
• 作者: Andrew Cropper;Sebastijan Dumancic;Stephen Muggleton

Complete Bottom-Up Predicate Invention in Meta-Interpretive Learning

元解释学习中完整的自下而上谓词发明

• DOI: 10.24963/ijcai.2020/320
• 发表时间: 2020
• 作者: Hocquette C

Learning higher-order logic programs

• DOI: 10.1007/s10994-019-05862-7
• 发表时间: 2019-12-03
• 期刊: MACHINE LEARNING
• 影响因子: 7.5
• 作者: Cropper, Andrew;Morel, Rolf;Muggleton, Stephen
• 通讯作者: Muggleton, Stephen

Ultra-Strong Machine Learning: comprehensibility of programs learned with ILP

超强机器学习：通过 ILP 学习的程序的可理解性

• DOI: 10.1007/s10994-018-5707-3
• 发表时间: 2018
• 期刊: Machine Learning
• 影响因子: 7.5
• 作者: Muggleton S

Learning efficient logic programs

学习高效的逻辑程序

• DOI: 10.1007/s10994-018-5712-6
• 发表时间: 2018
• 期刊: Machine Learning
• 影响因子: 7.5
• 作者: Cropper A