He has since then been active there, except for a sabbatical period (1986-1988) when he hold a Senior Scientist position at Microelectronics and Computer Technology Corporation in Austin, Texas.
In the mid 1980s he switched interests into machine learning methods, mostly artificial neural networks (ANN), for pattern recognition and image processing as well as algorithms for difficult optimization problems.
With regard to pattern recognition, he early on contributed with a learning algorithm for ANN, the mean field method to speed up the Boltzmann machine.
Since the early 1990s he has mostly been active within the interface between physics and biology with focus on biomarkers for serious diseases as well as gene regulation – the latter often with a network perspective approached with mechanistic methods.
Peterson is keen on whenever possible confronting his computational results with the real world and his activities are often performed imbedded within experimental groups.