Advertisement

Reverse Engineering Creativity into Interpretable Neural Networks

  • Marilena OitaEmail author
Conference paper
Part of the Lecture Notes in Networks and Systems book series (LNNS, volume 70)

Abstract

In the field of AI the ultimate goal is to achieve generic intelligence, also called “true AI”, but which depends on the successful enablement of imagination and creativity in artificial agents. To address this problem, this paper presents a novel deep learning framework for creativity, called INNGenuity. Pursuing an interdisciplinary implementation of creativity conditions, INNGenuity aims at the resolution of the various flaws of current AI learning architectures, which stem from the opacity of their models. Inspired by the neuroanatomy of the brain during creative cognition, the proposed framework’s hybrid architecture blends both symbolic and connectionist AI, inline with Minsky’s “society of mind”. At its core, semantic gates are designed to facilitate an input/output flow of semantic structures and enable the usage of aligning mechanisms between neural activation clusters and semantic graphs. Having as goal alignment maximization, such a system would enable interpretability through the creation of labeled patterns of computation, and propose unaligned but relevant computation patterns as novel and useful, therefore creative.

Keywords

Creativity Neural networks Imagination Semantic networks Knowledge Interpretability Neural architecture 

References

  1. 1.
    Weisberg, R.W.: The creative mind versus the creative computer. Behav. Brain Sci. 17(3), 555–557 (1994)CrossRefGoogle Scholar
  2. 2.
    Runco, M.A., Jaeger, G.J.: The standard definition of creativity. Creativity Res. J. (2012)Google Scholar
  3. 3.
    Schmidhuber, J.: Driven by compression progress: a simple principle explains essential aspects of subjective beauty, novelty, surprise, interestingness, attention, curiosity, creativity, art, science, music, jokes. CoRR (2008)Google Scholar
  4. 4.
    Terry, D.: Creativity, cognition, and knowledge: an interaction (2002)Google Scholar
  5. 5.
    Minsky, M.: The Society of Mind. Simon & Schuster Inc, New York (1986)Google Scholar
  6. 6.
    Jung, R.E., Segall, J.M., Bockholt, H., et al.: Neuroanatomy of creativity. Hum, Brain Mapp (2010)Google Scholar
  7. 7.
    Georgiev, G.V., Georgiev, D.D.: Enhancing user creativity semantic measures for idea generation: Knowl.-Based Syst. 151, 1–15 (2018)CrossRefGoogle Scholar
  8. 8.
    Wong, C., Gesmundo, A.: Transfer learning to learn with multitask neural model search. CoRR (2017)Google Scholar
  9. 9.
    Singh, P.: Examining the society of mind. Comput. Informat. 22, 521–543 (2004)MathSciNetzbMATHGoogle Scholar
  10. 10.
    Kaufman, S.B., Gabora, L.: Evolutionary approaches to creativity. Camb. Handb. Creativity, 279–300 (2011)Google Scholar
  11. 11.
    Baas, M., Nijstad, B.A., De Dreu, C.K.W.: (ed.): The cognitive, emotional and neural correlates of creativity. Front. Hum. Neurosci. 9, 275 (2015)Google Scholar
  12. 12.
    Fodor, J.A., Pylyshyn, Z.W.: Minds Without Meaning: An Essay on the Content of Concepts. The MIT Press, Cambridge (2015)Google Scholar
  13. 13.
    Karmiloff-Smith, A.: Is creativity domain specific or domain general? Cases from normal and abnormal phenotypes. Artif. Intell. Simul. Behav. Q. 85 (1993) (T.H. Dartnall)Google Scholar
  14. 14.
    Karmiloff-Smith, A.: Digest of beyond modularity. Behav. Brain Sci. 17(4) (1994)Google Scholar
  15. 15.
    Thagard, P.: Mind: Introduction to Cognitive Science. The MIT Press, Cambridge (1996)Google Scholar
  16. 16.
    Dawkins, R.: The Selfish Gene. 1941-(1989)Google Scholar
  17. 17.
    Singer, I.: Modes of Creativity: Philosophical Perspectives (2013)Google Scholar
  18. 18.
    Reboul, A.C.: Why language really is not a communication system: a cognitive view of language evolution. Front. Psychol. pp. 14–34 (2015)Google Scholar
  19. 19.
    Sperber, D., Wilson, D.: Relevance: Communication and Cognition. Basil Blackwel, Oxford (1995)Google Scholar
  20. 20.
    Holdgraf, C.R., Rieger, J.W., Michelli, C., Martin, S., Knight, R.T., Theunissen, F.E.: Encoding and decoding models in cognitive electrophysiology. Front. Syst. Neurosci. 11, 61 (2017)CrossRefGoogle Scholar
  21. 21.
    Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, L., Polosukhin, I.: Attention is all you need. CoRR, abs/1706.03762 (2017)Google Scholar
  22. 22.
    Bressler, S.L., Menon, V.: Large-scale brain networks in cognition: emerging methods and principles. Trends Cogn, Sci (2010)Google Scholar
  23. 23.
    Sporns, O., Honey C.J., Kotter, R.: Identification and classification of hubs in brain networks. PLoS ONE (2007)Google Scholar
  24. 24.
    Bassett, D.S., Bullmore, E.: Small-world brain networks. Neuroscientist 12, 512–523 (2006)CrossRefGoogle Scholar
  25. 25.
    Buckner, R.L., Andrews-Hanna, J.R., Schacter, D.L.: The brain’s default network: anatomy, function, and relevance to disease. Ann. N.Y. Acad. Sci. 1124 (2008)Google Scholar
  26. 26.
    Pascanu, R., Weber, T., Racanière, S., Reichert, D.P., Buesing, L., Guez, A., Rezende, D.J., Badia, A.P., Vinyals, O., Heess, N., Li, Y., Battaglia, P., Silver, D., Wierstra, D.: Imagination-augmented agents for deep reinforcement learning. CoRR (2017)Google Scholar
  27. 27.
    Hummel, J., Holyoak, K.: A symbolic-connectionist theory of relational inference and generalization. 110, 220–264 (2003)Google Scholar
  28. 28.
    Hummel, J.E., Holyoak, K.J.: Distributed representations of structure: a theory of analogical access and mapping. 104, 427–466 (1997)Google Scholar
  29. 29.
    Hochreiter, S., Schmidhuber, J.: Long short-term memory. Neural Comput. 9(8), 1735–1780 (1997)CrossRefGoogle Scholar
  30. 30.
    Mills, G.J., Healey, P.G.T.: Semantic negotiation in dialogue: the mechanisms of alignment (2008)Google Scholar
  31. 31.
    Campbell, D.T.: Blind variation and selective retention in creative thought as in other knowledge processes. Psychol. Rev. 67 (1960)CrossRefGoogle Scholar
  32. 32.
    Simonton, D.K.: Creative problem solving as sequential bvsr: exploration (total ignorance) versus elimination (informed guess). Thinking Skills Creativity 8 (2013)CrossRefGoogle Scholar
  33. 33.
    Peterson, J.C., Soulos, P., Nematzadeh, A., Griffiths, T.L.: Learning hierarchical visual representations in deep neural networks using hierarchical linguistic labels. CoRR, abs/1805.07647 (2018)Google Scholar
  34. 34.
    Jung, R., Mead, B., Carrasco, J., Flores, R.: The structure of creative cognition in the human brain. Front. Hum. Neurosci. 7, 330 (2013)Google Scholar
  35. 35.
    Bengio, Y., Courville, A., Vincent, P.: Representation learning: a review and new perspectives. IEEE Trans. Pattern Anal. Mach. Intell. 35(8), 1798–1828 (2013)CrossRefGoogle Scholar
  36. 36.
    Kenneth McGarry, S.W., MacIntyre, J.: Hybrid neural systems: from simple coupling to fully integrated neural networks. Neural Comput. Surv., pp. 62–93 (1999)Google Scholar
  37. 37.
    van Steenkiste, S., Chang, M., Greff, K., Schmidhuber, J.: Relational neural expectation maximization: unsupervised discovery of objects and their interactions. CoRR (2018)Google Scholar
  38. 38.
    Wang, X., Ye, Y., Gupta, A.: Zero-shot recognition via semantic embeddings and knowledge graphs. CoRR (2018)Google Scholar
  39. 39.
    Alvarez-Melis, D., Jaakkola, T.S.: A causal framework for explaining the predictions of black-box sequence-to-sequence models. CoRR, abs/1707.01943 (2017)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  1. 1.The Swiss AI Lab IDSIA, SUPSI, USILuganoSwitzerland

Personalised recommendations