Hypothesis Testing via Artificial Intelligence: Generating Physically Interpretable Models of Scientific Data with Machine Learning

Executive Summary

We will enable hypothesis testing via artificial intelligence by rapidly generating and testing hypotheses with machine learning and extracting physical insights and guiding the scientific process. This revolutionary approach will accelerate design-build-test iterations with application across the Department of Energy's and National Nuclear Security Administration's mission space, including stockpile stewardship, advanced manufacturing, and weather pattern prediction.

Publications, Presentations, and Patents

Petersen, Brenden K., Mikel Landajuela, Terrell N. Mundhenk, Claudio Prata Santiago, Soo Kyung Kim, and Joanne Taery Kim. "Deep symbolic regression: Recovering mathematical expressions from data via risk-seeking policy gradients." In International Conference on Learning Representations. 2021.

Landajuela, Mikel, Brenden K. Petersen, Sookyung Kim, Claudio P. Santiago, Ruben Glatt, Nathan Mundhenk, Jacob F. Pettit, and Daniel Faissol. "Discovering symbolic policies with deep reinforcement learning." In International Conference on Machine Learning. 2021.

Mundhenk, Nathan T., Mikel Landajuela, Ruben Glatt, Claudio P. Santiago, Daniel M. Faissol, and Brenden K. Petersen. Symbolic Regression via Deep Reinforcement Learning Enhanced Genetic Programming Seeding. In 2021 Conference on Neural Information Systems Processing. 2021.

Kim, Joanne T., Mikel Landajuela, and Brenden K. Petersen. "Distilling Wikipedia mathematical knowledge into neural network models." In Math-AI Workshop at ICML. 2021.

Landajuela, Mikel, Brenden K. Petersen, Soo K. Kim, Claudio P. Santiago, Ruben Glatt, T. Nathan Mundhenk, Jacob F. Pettit, and Daniel M. Faissol. "Improving exploration in policy gradient search: Application to symbolic optimization." In Math-AI Workshop at ICML. 2021.

Petersen, Brenden K., Claudio Santiago, and Mikel Landajuela. "Incorporating domain knowledge into neural-guided search via in situ priors and constraints." In 8th ICML Workshop on Automated Machine Learning. 2021.

Pettit, Jacob F., Brenden K. Petersen, F. L. Silva, Dale B. Larie, R. C. Cockrell, Gary An, and Daniel M. Faissol. Learning Sparse Symbolic Policies for Sepsis Treatment. In ICML Workshop for Interpretable Machine Learning in Healthcare. 2021.