Theory's Impact
Greg Yang discusses the potential impact of theory in the field of deep learning, emphasizing the importance of looking in the right places for the right problems to solve. He highlights the asymmetry between attackers and encryption applyers in cryptography as an example where theory excels in providing security guarantees. Daniel Bashir adds to the conversation, noting the financial and environmental costs of training massive models, making theory's application even more crucial in such high-stakes scenarios.In this clip
From this podcast
The Gradient
Greg Yang on Communicating Research, Tensor Programs, and µTransfer
Related Questions