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.
AlnajjarF.ZinI. B. M.MuraseK. (2009). A hierarchical autonomous robot controller for learning and memory: Adaptation in a dynamic environment. Adaptive Behavior, 17(3), 179–196.
2.
AnsB.RoussetS. (1997). Avoiding catastrophic forgetting by coupling two reverberating neural networks. Comptes Rendus de l’Académie des Sciences – Series {III} – Sciences de la Vie, 320(12), 989–997.
3.
BragagliaS.RayO. (2014). Nonmonotonic learning in large biological networks. In Proceedings of the 24th international conference on inductive logic programming.
4.
BrinS.PageL. (1998). The anatomy of a large-scale hypertextual web search engine. Computer Networks and ISDN Systems, 30(1–7), 107–117.
5.
CarlsonA.BetteridgeJ.KisielB.SettlesB.HruschkaE. R.Jr.MitchellT. M. (2010). Toward an architecture for never-ending language learning. In Proceedings of the 24th conference on artificial intelligence (pp. 1306–1313).
6.
CarpenterG.GrossbergS. (1988). The art of adaptive pattern recognition by a self-organizing neural network. Computer, 21(3), 77–88.
7.
Della SalaS. (2010). Forgetting. Hove: Psychology Press.
EatonE.RuvoloP. L. (2013). ELLA: An efficient lifelong learning algorithm. In ICML (Vol. 28, pp. 507–515).
10.
EiterT.WangK. (2008). Semantic forgetting in answer set programming. Artificial Intelligence, 172(14), 1644–1672.
11.
ErbasM. D.WinfieldA. F. T.BullL. (2014). Embodied imitation-enhanced reinforcement learning in multi-agent systems. Adaptive Behaviour, 22(1), 31–50.
12.
ErdemE.FerrarisP. (2007). Forgetting actions in domain descriptions. In Proceedings of the 22nd AAAI conference on artificial intelligence (pp. 409–414).
13.
Ferri-RamírezC.Hernández-OralloJ.Ramírez-QuintanaM. J. (2001). Incremental learning of functional logic programs. In Functional and logic programming (pp. 233–247).
14.
FrenchR. M. (1997). Pseudo-recurrent connectionist networks: An approach to the “sensitivity-stability” dilemma. Connection Science, 9, 353–379.
15.
GamaJ.ŽliobaitėI.BifetA.PechenizkiyM.BouchachiaA. (2014). A survey on concept drift adaptation. ACM Computing Surveys, 46(4), 44.
16.
GrossbergS. (2013). Adaptive resonance theory: How a brain learns to consciously attend, learn, and recognize a changing world. Neural Networks, 37, 1–47.
17.
HebbD. (1949). The organization of behavior. New York, NY: Wiley.
18.
HendersonR. (2014). Cumulative learning in the lambda calculus. Unpublished doctoral dissertation, Imperial College London, London, UK.
19.
Hernández-OralloJ. (2000). Constructive reinforcement learning. International Journal of Intelligent Systems, 15(3), 241–264.
20.
Hernández-OralloJ.García-VareaI. (2000). Explanatory and creative alternatives to the MDL priciple. Foundations of Science, 5(2), 185–207.
21.
HourdakisE.TrahaniasP. (2012). Computational modeling of observational learning inspired by the cortical underpinnings of human primates. Adaptive Behavior, 20(4), 237–256.
22.
KitzelmannE. (2010). Inductive programming: A survey of program synthesis techniques. In 3rd AAIP workshop (Vol. 5812).
23.
KleinbergJ. M. (1999). Authoritative sources in a hyperlinked environment. Journal of the ACM, 46(5), 604–632.
24.
LangJ.LiberatoreP. (2003). Propositional independence: Formula-variable independence and forgetting. Journal of Artificial Intelligence Research, 18, 391–443.
25.
LangJ.MarquisP. (2010). Reasoning under inconsistency: A forgetting-based approach. Artificial Intelligence, 174(12–13), 799–823.
26.
LempelR.MoranS. (2000). The stochastic approach for link-structure analysis (SALSA) and the TKC effect. Computer Networks, 33(1–6), 387–401.
27.
LeuG.CurtisN. J.AbbassH. A. (2014). Society of mind cognitive agent architecture applied to drivers adapting in a traffic context. Adaptive Behaviour, 22(2), 123–145.
28.
LiM.VitányiP. M. (2008). An introduction to Kolmogorov complexity and its applications. (3rd ed.). New York, NY: Springer Publishing Company.
29.
LinF.ReiterR. (1994). Forget it! In Proceedings of the AAAI fall symposium on relevance (pp. 154–159).
30.
LiuY.WenX. (2011). On the progression of knowledge in the situation calculus. In Proceedings of the international joint conference on artificial intelligence (Vol. 22, p. 976).
31.
Martínez-PlumedF.FerriC.Hernández-OralloJ.Ramírez-QuintanaM. J. (2013). Learning with configurable operators and RL-based heuristics. In New frontiers in mining complex patterns (pp. 1–16).
32.
Martínez-PlumedF.FerriC.Hernández-OralloJ.Ramírez-QuintanaM. J. (2014). A knowledge growth and consolidation framework for lifelong machine learning systems. In ICMLA (pp. 111–116).
33.
Martínez-PlumedF.FerriC.Hernández-OralloJ.Ramírez-QuintanaM. J. (2015). Forgetting and consolidation for incremental and cumulative knowledge acquisition systems. CoRR, abs/1502.05615.
34.
MuggletonS. H. (1995). Inverse entailment and Progol. New Generation Computing, 13(3–4), 245–286.
35.
MuggletonS. H.BainM.Hayes-MichieJ.MichieD. (1989). An experimental comparison of human and machine learning formalisms. In Proceedings of 6th international workshop on machine learning (pp. 113–118).
36.
MuggletonS. H.De RaedtL. (1994). Inductive logic programming: Theory and methods. The Journal of Logic Programming, 19, 629–679.
37.
NuxollA.TecuciD.HoW. C.WangN. (2010). Comparing forgetting algorithms for artificial episodic memory systems. In Proceedings of the symposium on human memory for artificial agents (pp. 14–20).
38.
QuirogaR. Q. (2012). Concept cells: The building blocks of declarative memory functions. Nature Reviews Neuroscience, 13, 587–597.
39.
RaducanuB.VitriàJ. (2008). Learning to learn: From smart machines to intelligent machines. Pattern Recognition Letters, 29(8), 1024–1032.
40.
RendellL. A. (1995). Lessons from theory revision applied to constructive induction. In ICML (p. 185).
41.
RissanenJ. (1999). Hypothesis selection and testing by the MDL principle. The Computer Journal, 42(4), 260–269.
42.
ShannonC. E. (1948). A mathematical theory of communication. Bell System Technical Journal, 28(4), 656–715.
43.
ShastriL. (2001). Biological grounding of recruitment learning and vicinal algorithms in long-term potentiation. In Emergent neural computational architectures based on neuroscience (Vol. 2036, pp. 348–367).
44.
SuK.SattarA.LvG.ZhangY. (2009). Variable forgetting in reasoning about knowledge. Journal of Artificial Intelligence Research, 35(2), 677.
45.
ThrunS. (1996). Is learning the n-th thing any easier than learning the first? In Advances in neural information processing systems (Vol. 8, pp. 640–646).
46.
WallaceC. (2005). Statistical and inductive inference by minimum message length. New York, NY: Springer.
47.
WallaceC. S.BoultonD. M. (1968). An information measure for classification. The Computer Journal, 11(2), 185–194.
48.
WallaceC. S.DoweD. L. (1999). Refinements of MDL and MML coding. Computer Journal, 42(4), 330–337.
49.
WangY.ZhangY.ZhouY.ZhangM. (2012). Forgetting in logic programs under strong equivalence. In Proceedings of the thirteenth international conference on principles of knowledge representation and reasoning (pp. 643–647).
50.
WangZ.WangK.ToporR.PanJ. (2010). Forgetting for knowledge bases in DL-Lite. Annals of Mathematics and Artificial Intelligence, 58(1–2), 117–151.
51.
WixtedJ. T. (2004). The psychology and neuroscience of forgetting. Annual Review of Psychology, 55, 235–269.
52.
WoodR.BaxterP.BelpaemeT. (2012). A review of long-term memory in natural and synthetic systems. Adaptive Behavior, 20(2), 81–103.
53.
ZhangY.FooN. Y. (2006). Solving logic program conflict through strong and weak forgettings. Artificial Intelligence, 170(8–9), 739–778.
