人们预计,通过集成表示学习和复杂推理的系统的出现,将实现向真正的人工智能的进展(LeCun et al. 2015)。针对这一预测,已经进行了在人工神经网络中实现冯·诺依曼计算机的符号推理的研究(Graves et al. 2016; Graves et al. 2014; Reed et al. 2015)。然而,这些研究在实现神经-符号积分方面有许多局限性(Jaeger. 2016年)。在这里,我们提出了一个新的学习范式:学习解决程序(LSP),学习解决复杂问题的程序。这不仅仅是通过学习输入-输出数据来实现的,而是通过通过求解过程的学习算法来实现的,该求解过程获得作为给定输入问题的任务序列的输出。LSP神经网络系统不仅学习简单的加法和乘法问题,而且学习复杂问题的算法,如复杂的算术表达式,排序和汉诺塔。为了实现这一点,LSP神经网络结构由递归组合的深度神经网络和长短期记忆组成。通过实验,我们证明了LSP的效率和可扩展性及其作为一种复杂推理机制的有效性。
英文摘要:
It is expected that progress toward true artificial intelligence will be achieved through the emergence of a system that integrates representation learning and complex reasoning (LeCun et al. 2015). In response to this prediction, research has been conducted on implementing the symbolic reasoning of a von Neumann computer in an artificial neural network (Graves et al. 2016; Graves et al. 2014; Reed et al. 2015). However, these studies have many limitations in realizing neural-symbolic integration (Jaeger. 2016). Here, we present a new learning paradigm: a learning solving procedure (LSP) that learns the procedure for solving complex problems. This is not accomplished merely by learning input-output data, but by learning algorithms through a solving procedure that obtains the output as a sequence of tasks for a given input problem. The LSP neural network system not only learns simple problems of addition and multiplication, but also the algorithms of complicated problems, such as complex arithmetic expression, sorting, and Hanoi Tower. To realize this, the LSP neural network structure consists of a deep neural network and long short-term memory, which are recursively combined. Through experimentation, we demonstrate the efficiency and scalability of LSP and its validity as a mechanism of complex reasoning.
Solving tricky quantum optics problems with assistance from (artificial) intelligence
Indian Institute of Technology, Kanpur(印度理工学院,坎pur);MPA-Quantum, Los Alamos National Laboratory(MPA-量子,洛斯阿拉莫斯国家实验室);Johannes Gutenberg-Universität Mainz(美因茨约翰尼斯·古滕贝格大学);Helmholtz Institute Mainz(马因茨海德堡研究所);GSI Helmholtzzentrum für Schwerionenforschung GmbH(重离子研究海德堡中心 GmbH);Department of Physics, University of California, Berkeley(加州大学伯克利分校物理系)
An Artificial Intelligence driven Learning Analytics Method to Examine the Collaborative Problem solving Process from a Complex Adaptive Systems Perspective
College of Education, Zhejiang University(浙江大学教育学院);UCL Knowledge Lab, University College London(伦敦大学学院UCL知识实验室)
A Heuristic Method for Solving the Problem of Partitioning Graphs with Supply and Demand
Institute of Physics, University of Belgrade(贝尔格莱德大学物理研究所);Qatar Environment and Energy Research Institute (QEERI)(卡塔尔环境与能源研究所);Institute of Information Systems, University of Hamburg(汉堡大学信息系统研究所)
Solving Generalized Grouping Problems in Cellular Manufacturing Systems Using a Network Flow Model
Department of Mechanical Engineering, Khulna University of Engineering & Technology(库尔纳工程技术大学机械工程系);Department of Industrial Engineering and Management, Khulna University of Engineering & Technology(库尔纳工程技术大学工业工程与管理系);Thomas Lord Department of Computer Science, Viterbi School of Engineering, University of Southern California(南加州大学维特比工程学院托马斯·洛德计算机科学系);Department of Computer Science, American International University-Bangladesh(美国国际大学-孟加拉国计算机科学系)
The Foundations of Computational Management: A Systematic Approach to Task Automation for the Integration of Artificial Intelligence into Existing Workflows
Vivenxia Group LLC(Vivenxia集团有限责任公司);Centre for Digital Therapeutics(数字治疗中心);University of Southern California(南加州大学)