Imposing constraints on the output of a Deep Neural Net is one way to improve the quality of its predictions while loosening the requirements for labeled training data. Such constraints are usually imposed as soft constraints by adding new terms to the loss function that is minimized during training. An alternative is to impose them as hard constraints, which has a number of theoretical benefit...

I propose and defend a novel view called “de se consequentialism,” which is noteworthy for two reasons. First, it demonstrates—contra Doug Portmore, Mark Schroeder, Campbell Brown, Michael Smith, among others—that agent-neutral consequentialism consistent with agent-centered constraints. Second, clarifies the nature of constraints, thereby meriting attention from even dedicated nonconsequential...

Abstract What is the philosophical role of an historical account how someone, or some people, came to believe value as they do? I consider proposals, due Bernard Williams and David Wiggins, according which such might either vindicate subvert our believing valuing we do. suggest reasons for scepticism about those at least when construed providing a fundamental means assessing cases valuing. The ...