AI software writes, and rewrites, its own code, getting smarter as it does

Gamalon uses a technique that it calls Bayesian program synthesis to build algorithms capable of learning from fewer examples. "Probabilistic programming will make machine learning much easier for researchers and practitioners," Lake says. In theory, Gamalon's approach could make it a lot easier for someone to build and refine a machine-learning model, too. The company's first products use Bayesian program synthesis to recognize concepts in text.