Knowledge acquisition with forgetting: an incremental and developmental setting

Identifying the balance between remembering and forgetting is the key to abstraction in the human brain and, therefore, the creation of memories and knowledge. We present an incremental, lifelong view of knowledge acquisition which tries to improve task after task by determining what to keep, consolidate and forget, overcoming the stability–plasticity dilemma. Our framework can combine any rule-based inductive engine (which learns new rules) with a deductive engine (which derives a coverage graph for all rules) and integrate them into a lifelong learner. We rate rules by introducing several metrics through the first adaptation, to our knowledge, of the minimum message length (MML) principle to a coverage graph, a hierarchical assessment structure which handles evidence and rules in a unified way. The metrics are used to forget some of the worst rules and also to consolidate those selected rules that are promoted to the knowledge base. This mechanism is also mirrored by a demotion system. We evaluate the framework with a series of tasks in a chess rule learning domain.

[1]  Ming Li,et al.  An Introduction to Kolmogorov Complexity and Its Applications , 1997, Texts in Computer Science.

[2]  Stephen Grossberg,et al.  The ART of adaptive pattern recognition by a self-organizing neural network , 1988, Computer.

[3]  C. S. Wallace,et al.  Statistical and Inductive Inference by Minimum Message Length (Information Science and Statistics) , 2005 .

[4]  Stephen Muggleton,et al.  An Experimental Comparison of Human and Machine Learning Formalisms , 1989, ML.

[5]  Jeff Z. Pan,et al.  Forgetting for knowledge bases in DL-Lite , 2010, Annals of Mathematics and Artificial Intelligence.

[6]  J. Ramon,et al.  Nonmonotonic Learning in Large Biological Networks , 2016 .

[7]  José Hernández-Orallo,et al.  A Knowledge Growth and Consolidation Framework for Lifelong Machine Learning Systems , 2014, 2014 13th International Conference on Machine Learning and Applications.

[8]  Yan Zhang,et al.  Variable Forgetting in Reasoning about Knowledge , 2009, J. Artif. Intell. Res..

[9]  J. Hernández-Orallo.  Constructive reinforcement learning , 2000 .

[10]  Stephen Grossberg,et al.  Adaptive Resonance Theory: How a brain learns to consciously attend, learn, and recognize a changing world , 2013, Neural Networks.

[11]  Bogdan Raducanu,et al.  Learning to learn: From smart machines to intelligent machines , 2008, Pattern Recognit. Lett..

[12]  João Gama,et al.  A survey on concept drift adaptation , 2014, ACM Comput. Surv..

[13]  David L. Dowe,et al.  Refinements of MDL and MML Coding , 1999, Comput. J..

[14]  Emanuel Kitzelmann,et al.  Inductive Programming: A Survey of Program Synthesis Techniques , 2009, AAIP.

[15]  E. Capaldi,et al.  The organization of behavior. , 1992, Journal of applied behavior analysis.

[16]  Peter Grünwald,et al.  Invited review of the book Statistical and Inductive Inference by Minimum Message Length , 2006 .

[17]  Kazuyuki Murase,et al.  A Hierarchical Autonomous Robot Controller for Learning and Memory: Adaptation in a Dynamic Environment , 2009, Adapt. Behav..

[18]  L’oubli catastrophique it,et al.  Avoiding catastrophic forgetting by coupling two reverberating neural networks , 2004 .

[19]  José Hernández-Orallo,et al.  Forgetting and consolidation for incremental and cumulative knowledge acquisition systems , 2015, ArXiv.

[20]  J. Wixted.  The psychology and neuroscience of forgetting. , 2004, Annual review of psychology.

[21]  C. S. Wallace,et al.  An Information Measure for Classification , 1968, Comput. J..

[22]  Sang Joon Kim,et al.  A Mathematical Theory of Communication , 2006 .

[23]  Norman Y. Foo,et al.  Solving Logic Program Conflict through Strong and Weak Forgettings , 2005, IJCAI.

[24]  Luc De Raedt,et al.  Inductive Logic Programming: Theory and Methods , 1994, J. Log. Program..

[25]  Claude E. Shannon,et al.  Communication theory of secrecy systems , 1949, Bell Syst. Tech. J..

[26]  J. Hernández-Orallo,et al.  Explanatory and Creative Alternatives to the MDL priciple , 2000 .

[27]  Robert John Henderson,et al.  Cumulative learning in the lambda calculus , 2013 .

[28]  Sergey Brin,et al.  The Anatomy of a Large-Scale Hypertextual Web Search Engine , 1998, Comput. Networks.

[29]  Wan Ching Ho,et al.  Comparing Forgetting A lgorithms for A rtificial Episodic Memory Systems , 2010 .

[30]  José Hernández-Orallo,et al.  Incremental Learning of Functional Logic Programs , 2001, FLOPS.

[31]  Yi Zhou,et al.  Forgetting in Logic Programs under Strong Equivalence , 2012, KR.

[32]  Panos E. Trahanias,et al.  Computational modeling of observational learning inspired by the cortical underpinnings of human primates , 2012, Adapt. Behav..

[33]  Yongmei Liu,et al.  On the Progression of Knowledge in the Situation Calculus , 2011, IJCAI.

[34]  Jorma Rissanen,et al.  Hypothesis Selection and Testing by the MDL Principle , 1999, Comput. J..

[35]  R. Reiter,et al.  Forget It ! , 1994 .

[36]  Robert M. French,et al.  Pseudo-recurrent Connectionist Networks: An Approach to the 'Sensitivity-Stability' Dilemma , 1997, Connect. Sci..

[37]  Kewen Wang,et al.  Semantic forgetting in answer set programming , 2008, Artif. Intell..

[38]  R. Quiroga.  Concept cells: the building blocks of declarative memory functions , 2012, Nature Reviews Neuroscience.

[39]  Stephen Muggleton,et al.  Inverse entailment and progol , 1995, New Generation Computing.

[40]  Pierre Marquis,et al.  Reasoning under inconsistency: A forgetting-based approach , 2010, Artif. Intell..

[41]  Hussein A. Abbass,et al.  Society of Mind cognitive agent architecture applied to drivers adapting in a traffic context , 2014, Adapt. Behav..

[42]  José Hernández-Orallo,et al.  Learning with Configurable Operators and RL-Based Heuristics , 2012, NFMCP.

[43]  Shlomo Moran,et al.  The stochastic approach for link-structure analysis (SALSA) and the TKC effect , 2000, Comput. Networks.

[44]  Lokendra Shastri,et al.  Biological Grounding of Recruitment Learning and Vicinal Algorithms in Long-Term Potentiation , 2001, Emergent Neural Computational Architectures Based on Neuroscience.

[45]  Sebastian Thrun,et al.  Is Learning The n-th Thing Any Easier Than Learning The First? , 1995, NIPS.

[46]  Eric Eaton,et al.  ELLA: An Efficient Lifelong Learning Algorithm , 2013, ICML.

[47]  Esra Erdem,et al.  Forgetting Actions in Domain Descriptions , 2007, AAAI.

[48]  Estevam R. Hruschka,et al.  Toward an Architecture for Never-Ending Language Learning , 2010, AAAI.

[49]  Tony Belpaeme,et al.  A review of long-term memory in natural and synthetic systems , 2012, Adapt. Behav..

[50]  Larry A. Rendell,et al.  Lessons from Theory Revision Applied to Constructive Induction , 1995, ICML.

[51]  Larry Bull,et al.  Embodied imitation-enhanced reinforcement learning in multi-agent systems , 2014, Adapt. Behav..

[52]  Faith T. Fitzgerald,et al.  Forgetting , 2011, Journal of General Internal Medicine.

[53]  Yan Zhang,et al.  Knowledge forgetting: Properties and applications , 2009, Artif. Intell..