Timezone: »

I (Still) Can't Believe It's Not Better: A workshop for “beautiful” ideas that "should" have worked
Aaron Schein · Melanie F. Pradier · Jessica Forde Jessica Forde · Stephanie Hyland · Francisco Ruiz

Mon Dec 13 05:00 AM -- 05:00 PM (PST) @ None
Event URL: https://i-cant-believe-its-not-better.github.io/neurips2021/ »

Beautiful ideas have shaped scientific progress throughout history. As Paul Dirac said, “If one is working from the point of view of getting beauty in one's equations, (…), one is on a sure line of progress.” However, beautiful ideas are often overlooked in a research environment that heavily emphasizes state-of-the-art (SOTA) results, where the worth of scientific works is defined by their immediate utility and quantitative superiority instead of their creativity, diversity, and elegance. This workshop will explore gaps between the form and function (or, the intrinsic and extrinsic value) of ideas in ML and AI research. We will explore that disconnect by asking researchers to submit their “beautiful” ideas that don’t (yet) “work”. We will ask them to explain why their idea has intrinsic value, and hypothesize why it hasn’t (yet) shown its extrinsic value. In doing so, we will create a space for researchers to help each other get their “beautiful” ideas “working”.

Author Information

Aaron Schein (Columbia University)
Melanie F. Pradier (Microsoft Research)
Jessica Forde Jessica Forde (Brown University)
Stephanie Hyland (Microsoft Research)
Francisco Ruiz (DeepMind)

More from the Same Authors