Glad to share my last work on NLI for Persian. You can read more about the techniques used, or finetune the model by yourself using the published notebooks. Also, the model is published on the HuggingFace hub to use in tasks like zero-shot classification.
https://t.co/8FsuvHvSxJ
I'm super excited to share all the things we've spent the past year+ working on at @elastic. The new @PyTorch NLP and ANN w/ HNSW in #Lucene 9 are just the very beginnings of what we're planning on bringing to the 8.x series!
https://t.co/qD30CnrfVP
New video!
⬜⬜⬜⬜⬜
🟩🟨🟩⬜⬜
🟩🟩🟩🟩🟩
It's an exploration for writing a Wordle solver, with the challenge of not using the official list of Wordle answers (except as a test set), which is really just an excuse for an information theory lesson.
https://t.co/Tyh1DTtbUm
Introducing #AlphaCode: a system that can compete at average human level in competitive coding competitions like @codeforces. An exciting leap in AI problem-solving capabilities, combining many advances in machine learning!
Read more: https://t.co/yaXfMWtEfe 1/
@svpino 1) single sample -> stochastic gradient descent
2) total samples -> batch gradient descent
3) more than one sample but fewer than entire dataset -> mini-batch gradient descent