Phd thesis01 Jul 2014
My work focused on learning recurring patterns in multimodal perception. For that purpose we developed cognitive systems that model the mechanisms providing such capabilities to infants.
More precisely, my thesis revolves around two main topics that are, on the one hand the ability of infants or robots to imitate and understand human behaviors, and on the other the acquisition of language. At the crossing of these topics, it studies the question of the how a developmental cognitive agent can discover a dictionary of primitive patterns from its multimodal perceptual flow. It specifies this problem and formulate its links with Quine’s indetermination of translation and blind source separation, as studied in acoustics.
The thesis sequentially studies four sub-problems and provide an experimental formulation of each of them. It then describes and tests computational models of agents solving these problems. They are particularly based on bag-of-words techniques, matrix factorization algorithms, and inverse reinforcement learning approaches. It first goes in depth into the three separate problems of learning primitive sounds, such as phonemes or words, learning primitive dance motions, and learning primitive objective that compose complex tasks. Finally it studies the problem of learning multimodal primitive patterns, which corresponds to solve simultaneously several of the aforementioned problems. It also details how the last problems models acoustic words grounding.