An Approach to Combining Explanation-based and Neural Learning Algorithms

Shavlik, Jude W.; Towell, Geoffrey G.

doi:10.1142/9789812812414_0004

Cited by 26 publications

(32 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Essentially the system used UL to greatly reduce problem depth. Compare earlier BP-based fine-tuning of NNs initialized by rules of propositional logic (Shavlik and Towell, 1989) (Sec. 5.6.1).…”

Section: : Ul-based History Compression Through a Deep Stack Of Rnnsmentioning

confidence: 98%

Deep learning in neural networks: An overview

2015

View full text Add to dashboard Cite

In recent years, deep artificial neural networks (including recurrent ones) have won numerous contests in pattern recognition and machine learning. This historical survey compactly summarises relevant work, much of it from the previous millennium. Shallow and deep learners are distinguished by the depth of their credit assignment paths, which are chains of possibly learnable, causal links between actions and effects. I review deep supervised learning (also recapitulating the history of backpropagation), unsupervised learning, reinforcement learning & evolutionary computation, and indirect search for short programs encoding deep and large networks.LATEX source: http://www.idsia.ch/˜juergen/DeepLearning8Oct2014.tex Complete BIBTEX file (888 kB): http://www.idsia.ch/˜juergen/deep.bib Preface This is the preprint of an invited Deep Learning (DL) overview. One of its goals is to assign credit to those who contributed to the present state of the art. I acknowledge the limitations of attempting to achieve this goal. The DL research community itself may be viewed as a continually evolving, deep network of scientists who have influenced each other in complex ways. Starting from recent DL results, I tried to trace back the origins of relevant ideas through the past half century and beyond, sometimes using "local search" to follow citations of citations backwards in time. Since not all DL publications properly acknowledge earlier relevant work, additional global search strategies were employed, aided by consulting numerous neural network experts. As a result, the present preprint mostly consists of references. Nevertheless, through an expert selection bias I may have missed important work. A related bias was surely introduced by my special familiarity with the work of my own DL research group in the past quarter-century. For these reasons, this work should be viewed as merely a snapshot of an ongoing credit assignment process. To help improve it, please do not hesitate to send corrections and suggestions to juergen@idsia.ch.

show abstract

Section: : Ul-based History Compression Through a Deep Stack Of Rnnsmentioning

confidence: 98%

Deep learning in neural networks: An overview

2015

View full text Add to dashboard Cite

show abstract

“…Hence, to incorporate the observer's advice, the agent's neural network must be updated. We use ideas from knowledge-based neural networks (Fu, 1989;Omlin & Giles, 1992;Shavlik & Towell, 1989) to directly install the advice into the agent. In one approach to knowledge-based neural networks, KBANN (Towell, Shavlik, & Noordewier, 1990;Towell & Shavlik, 1994), a set of propositional rules is re-represented as a neural network.…”

Section: A General Framework For Advice-takingmentioning

confidence: 99%

“…There has been a growing literature on automated "theory refinement" (Fu, 1989;Ginsberg, 1988;Ourston & Mooney, 1994;Pazzani & Kibler, 1992;Shavlik & Towell, 1989), and it is from this research perspective that our advice-taking work arose. Our new work differs by its novel emphasis on theory refinement in the context of multi-step problem solving in multi-actor worlds, as opposed to refinement of theories for categorization and diagnosis.…”

Section: Refining Prior Knowledgementioning

confidence: 99%

Creating advice-taking reinforcement learners

Maclin

Shavlik

1996

Mach Learn

Self Cite

View full text Add to dashboard Cite

Abstract. Learning from reinforcements is a promising approach for creating intelligent agents. However, reinforcement learning usually requires a large number of training episodes. We present and evaluate a design that addresses this shortcoming by allowing a connectionist Q-learner to accept advice given, at any time and in a natural manner, by an external observer. In our approach, the advice-giver watches the learner and occasionally makes suggestions, expressed as instructions in a simple imperative programming language_ Based on techniques from knowledge-based neural networks, we insert these programs directly into the agent's utility function. Subsequent reinforcement learning further integrates and refines the advice. We present empirical evidence that investigates several aspects of our approach and shows that, given good advice, a learner can achieve statistically significant gains in expected reward. A second experiment shows that advice improves the expected reward regardless of the stage of training at which it is given, while another study demonstrates that subsequent advice can result in further gains in reward. Finally, we present experimental results that indicate our method is more powerful than a naive technique for making use of advice.

show abstract

“…This complementarity naturally suggests that by properly integrating these methods, one could obtain a synergistic effect in which different strategies mutually support each other and compensate for each other's weaknesses. This hypothesis has been confirmed by the many multistrategy learning methods and systems that have been developed in the past several years (e.g., Bergadano & Giordana, 1990;Cox & Ram, 1991;Danyluk, 1987;DeRaedt & Bruynooghe, 1991;Flann & Dietterich, 1989;Genest, Matwin, & Plante, 1990;Hirsh, 1989;Lebowitz, 1986;Minton & Carbonell, 1987;Mooney & Ourston, 1991;Morik, 1993;Pazzani, 1988;Reich, 1991;Saitta & Botta, 1993;Shavlik & Towell, 1990;Tecuci & Kodratoff, 1990;Whitehall, 1990;Widmer, 1991;Wilkins, 1990).…”

Section: Introductionmentioning

confidence: 85%

Untitled

Tecuci

1993

Machine Learning

View full text Add to dashboard Cite

Abstract. This article describes a framework for the deep and dynamic integration of learning strategies. The framework is based on the idea that each single-strategy learning method is ultimately the result of certain elementary inferences (like deduction, analogy, abduction, generalization, specialization, abstraction, concretion, etc.). Consequently, instead of integrating learning strategies at a macro level, we propose to integrate the different inference types that generate individual learning strategies. The article presents a concept-learning and theory-revision method that was developed in this framework. It allows the system to learn from one or from several (positive and/or negative) examples, and to both generalize and specialize its knowledge base. The method integrates deeply and dynamically different learning strategies, depending on the relationship between the input information and the knowledge base. It also behaves as a single-strategy learning method whenever the applicability conditions of such a method are satisfied.

show abstract

An Approach to Combining Explanation-based and Neural Learning Algorithms

Cited by 26 publications

References 0 publications

Deep learning in neural networks: An overview

Deep learning in neural networks: An overview

Creating advice-taking reinforcement learners

Untitled

Contact Info

Product

Resources

About