Philip Thomas, UMass Amherst (IMAGE) University of Massachusetts Amherst Caption Philip Thomas at UMass Amherst, with colleagues there and at Stanford, says they say they hope that machine learning researchers will go on to develop new and more sophisticated algorithms using a new framework they developed based on reasoning probabilistically about safety, which can be used responsibly for applications where machine learning used to be considered too risky. They also call on others to conduct research in this space. Credit UMass Amherst Usage Restrictions None License Licensed content Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.