Separately from the main contribution of the paper, I'm also still a big fan of our extension of Ethan Perez's FiLM to the multi-modal case, which we call FiLM². In the age of the transformer, I think this sort of affine cross-modal interaction hasn't been explored enough [4/14]

"Clarity is the elimination of mental clutter. Agility is the elimination of physical clutter. Tranquility is the elimination of spiritual clutter." – James Clear

Do you want to learn about current happenings in ethics, bias, and fairness in natural language processing? Weekly meetings are on Tuesday at 6:00 PM EST. Discussion Group Name: Bias and Fairness in Natural Language Processing Please sign up (free) at lnkd.in/gViGRT2

The intersection if AI and neuroscience is a research field that can fundamentally reshape our understanding of consciousness. On the TDS Podcast, Jeremie Harris spoke with Georg Northoff The Royal Université d'Ottawa | University of Ottawa, who is at the forefront of this work. buff.ly/3cFYRsM

🌐 Data Science Interviews Includes a curated list of data science interview resources and types of questions. by Alexey Grigorev ds-interviews.org

Check out my recent blogpost on different language models trained and used in the biomedical and/or clinical domain and how their performances compare in measuring text similarity. #NLP #healthcare techblog.ezra.com/semantic-simil…

Join us on May 25th to hear from an exceptional group of multi-disciplinary builders and creators designing a more diverse, inclusive, and ethical AI/tech future! #RethinkAI Register ⬇️ bit.ly/32TjuNz

I couldn't resist. With contributions from Max Kleiman-Weiner

Skepticism is the heart of science, but needs to be informed skepticism. We shouldn't blindly follow experts, but should acknowledge our own biases and preconceptions when encountering new evidence. This video by Neil Halloran strikes a perfect balance: youtube.com/watch?v=R7FAAf…

Get your ticket now for Women in AI Ethics™'s summit "Rethink AI: Ethical AI Tech & Tools" 🗓️ Tue, May 25 🕗 8AM-4PM PDT Then tune in for our panel "Building Trustworthy AI" at 9.35AM with @chenaichair, Kathy Pham, @wiebketous & Temi Popo eventbrite.com/e/rethink-ai-e…

Checkout my recent blogpost #medium at Towards Data Science towardsdatascience.com/stochastic-par…

I also recommend his Tedx talk: m.youtube.com/watch?v=stxJMs…

Challenges and Opportunities in NLP Benchmarking Recent NLP models have outpaced the benchmarks to test for them. I provide an overview of challenges and opportunities in this blog post. ruder.io/nlp-benchmarki…

A Comprehensive Survey of Activation Functions Activation functions play a key role in the effectiveness of neural networks for deep learning. This new survey summarizes the most popular ones with insights useful for ML practitioners and students. arxiv.org/abs/2109.14545

We're excited to announce that Timnit Gebru @timnitGebru (@dair-community.social/bsky.social) will be doing the opening keynote for our annual summit! 🤩 #AIChangemakers Don't miss out on our phenomenal lineup of speakers. Register now ➡️ eventbrite.com/e/100-brillian…

DATA-CENTRISM AND ITS DISCONTENTS: Workshop #1, on “STOCHASTIC PARROTS” criticalai.org/2021/10/14/blo… via CriticalAIhttps://read.dukeupress.edu/critical-ai/

For those playing along at home, here's a "AI is sentient!" argument bingo card.

Wow! This is exciting! First ever course on Transformers by Stanford. Really looking forward to the release of all lectures. website: web.stanford.edu/class/cs25/ youtube: youtube.com/playlist?list=…

Neat visual explanations of several important topics in machine learning: logistic regression, ROC & AUC, precision & recall, random forest, decision trees, and more. These are great references to review before an ML interview. Check it out here: mlu-explain.github.io