August 26, 2020

[D] A question about GANs output variety

Hi, I am new to machine learning, and I am studying GANs. I built a simple vanilla GANs to learn just to generate a set of […]
August 26, 2020

[D] Generalization problem with large batch sizes, why are we using TPUs?

Multiple publications and blogs mentions that training with large batch sizes reduce the generalization and the model receives higher test error. TPUs are devices which allows […]
August 27, 2020

[D] Why ML conference reviews suck – A video analysis of the incentive structures behind publishing & reviewing.

https://youtu.be/VgqHitvEbR0 Machine Learning research is in dire straits as more people flood into the field and competent reviewers are scarce and overloaded. This video takes a […]
August 27, 2020

Andrej Karpathy releases concise GPT implementation. Why has he bothered to do this: doesn't he work for OpenAI, at least indirectly? [D] [N]

It's nice to see a concise implementation of GPT, in pytorch, as it is true Hugging Face's Transformer's is excellent, but it is quite difficult to […]