Talent Highlight: Daniel Elton
Co-author of: Applying machine learning techniques to predict the properties of energetic materials
Dan has been interested in biologically inspired AI since 2010, when he did research on neuroscience-inspired computer vision at Los Alamos National Laboratory. He has written about neuromorphic chips and machine learning topics on his blog. In 2016 he attended the “Ethics of AI” conference at NYU and in 2017 he found out about Mindfire from Joanna Bryson while attending the Envision Conference at Princeton University. After participating in Mindfire Mission-1 he organized an online journal club with other MindFire members to study the free energy principle of brain function and brainstorm ideas.
Dan grew up in the small town of Glens Falls, New York. He obtained a Ph.D. in physics in 2016 from Stony Brook University. His Ph.D. research focused on studying how water absorbs electromagnetic radiation using molecular dynamics simulations.
Dan is currently an assistant research scientist at the University of Maryland, College Park, where he does research on applications of artificial intelligence to molecular design and discovery. His research is highly interdisciplinary as it involves applying machine learning to predicting the chemical and physical properties of molecules and materials. Some of his work was recently published in Scientific Reports. In addition to using machine learning for property prediction he also works on using deep learning architectures such as generative adversarial networks to generate candidate molecules which are predicted to have desirable properties. There is currently a lot of excitement about the potential uses of deep learning and AI in drug discovery. Additionally, Dan has worked with natural language processing techniques such as word embeddings to try to extract information about chemicals from large bodies of text. One of the key applications of AI in the next few years will be assisting humans in extracting useful information from massive collections of scientific literature.
Dan is excited to apply his physics & neuroscience knowledge, machine learning expertise, and programming chops to the Mindfire mission!