scientific discovery Archives

Adaptive experiments: Machine learning can help scientific discovery

Written by

Cheng Soon Ong develops statistical machine learning methods and applies machine learning-guided design to biology.

Machine learning can help scientists design experiments. Scientific discovery relies on experiments that build our understanding of natural phenomena, and traditionally has been based on trial and error. Depending on the goal, different machine learning strategies can be used for adaptive experiments: active learning, maximising information gain, Bayesian optimisation, bandit approaches, and reinforcement learning. Cheng Soon Ong, machine learning scientist […]

Reframing understanding as a phase transition

Written by

When understanding comes to us, it can come suddenly. As we finally see how the different parts of a problem fall into place, it can bring about feelings of satisfaction, and perhaps even frustration that we did not discern the seemingly obvious solution beforehand. In their research, Dipl.-Phil. Elena Popova and Professor Valentin Popov at Technische Universität Berlin argue that […]