Hot take: a sentence starting with "In other words,..." is a sign that you wrote the previous sentence poorly and are just too lazy to re-write it clearly.
— Tatyana Deryugina (@TDeryugina) January 18, 2021
Hot take: a sentence starting with "In other words,..." is a sign that you wrote the previous sentence poorly and are just too lazy to re-write it clearly.
— Tatyana Deryugina (@TDeryugina) January 18, 2021
Indeed, the 2nd half is highly worthwhile - inspired a lot of our current work on GTN. Do yourself a favor: https://t.co/Q6YARJyuFV
— Awni Hannun (@awnihannun) January 17, 2021
The GTN framework and corresponding preprint are here:
Code: https://t.co/5QkN5v3RFD
Paper: https://t.co/ajaWaKDr2B https://t.co/CCorfDB7Ag
The 50 best public datasets for machine learning: https://t.co/yaAVYxzsMy
— MIT CSAIL (@MIT_CSAIL) January 17, 2021
(v/@dd_invest @Medium) pic.twitter.com/TzsAroenBo
Excluding sex might be an easy way to remove all pages with sexual content, but it will also exclude any page mentioning same sex couples, same sex families... More thoughts (and diverse teams) are needed when building those huge datasets that train big LM models! https://t.co/noXGRYCxW2
— Sylvain Gugger (@GuggerSylvain) January 17, 2021
Reproducible webscraping tip that saved me big time today: if you're scraping minor things, like a single table from a website, prefix the URL with an Internet Archive snapshot like this, bc the site and/or URL will inevitably change! pic.twitter.com/l4Hm25CQRI
— Andrew Heiss (@andrewheiss) January 16, 2021
GPU accelerated SHAP values in the latest version of XGBoost are awesome. Check them out. @RAPIDSai @nvidia https://t.co/JXeQB6Yhoh
— Bojan Tunguz (@tunguz) January 16, 2021
Machine Learning: The Great Stagnation https://t.co/tMQpoKv3xu
— /MachineLearning (@slashML) January 16, 2021
Introducing ToTTo, an open domain table-to-text generation dataset, a challenging benchmark for high-precision text generation, and a controlled text generation task that can be used to assess model “hallucination”. Read more and grab the code at https://t.co/9NaFcZvFo5
— Google AI (@GoogleAI) January 15, 2021
My talk on "Interpretable ML - A Brief History, State-of-the-Art and Challenges" is on YouTube: https://t.co/aNqKU9mh7C https://t.co/NdJc4EsXFb
— Christoph Molnar (@ChristophMolnar) January 15, 2021
In a new paper, our team uses unsupervised program synthesis to make sense of sensory sequences. This system is able to solve intelligence test problems zero-shot, without prior training on similar tasks: https://t.co/ifxANPWcIZ pic.twitter.com/HHWG3qEAg1
— DeepMind (@DeepMind) January 15, 2021
🚀Excited to release Robustness Gym, a new Python evaluation toolkit for evaluating the robustness of NLP models, as part of a collaboration between Stanford, Salesforce Research and UNC Chapel-Hill.
— Karan Goel (@krandiash) January 14, 2021
Paper: https://t.co/WjcfdE42x1
Code: https://t.co/1V4fAtJvcD
pip install!
First 2 lectures' slides/notes on machine learning systems design are up! Covering:
— Chip Huyen (@chipro) January 14, 2021
- Batch vs. online prediction, edge vs. cloud, offline vs. online learning
- Multiple-objective optimization
- When & when not to use ML
...
Feedback much appreciated!https://t.co/fZjtsKC1b4