From Anna Menacher: A timeline of the most important statistical ideas of the past 50 years https://t.co/nusBOYWwD6

— Andrew Gelman et al. (@StatModeling) June 16, 2022

Torch-TensorRT is now an official part of the PyTorch ecosystem and now available on PyTorch GitHub and Documentation. Torch-TensorRT is a TensorRT integration for PyTorch that accelerates inference up to 4x on NVIDIA GPUs with just a single line of code. https://t.co/cGd4g8FDeQ pic.twitter.com/AnnRr0SrN1

— PyTorch (@PyTorch) June 16, 2022

Again I'm starting to see comments in support of LMs learning meaning invoking the lived experiences of Blind people, from those who don't appear to have said lived experiences. Please stop.https://t.co/M73ykMMu72

— Emily M. Bender (@emilymbender) June 15, 2022

Efficient Decoder-free Object Detection with Transformers
abs: https://t.co/YW4QcqztiW

experiments on the MS COCO benchmark demonstrate that DFFT_SMALL outperforms DETR by 2.5% AP with 28% computation cost reduction and more than 10× fewer training epochs pic.twitter.com/ICOvqgA8xQ

— AK (@_akhaliq) June 15, 2022

Peripheral Vision Transformer
abs: https://t.co/c6R8BfNDPS

propose to incorporate peripheral position encoding to the multi-head self-attention layers to let the network learn to partition the visual field into diverse peripheral regions given training data pic.twitter.com/S78e7WXDKh

— AK (@_akhaliq) June 15, 2022

Pro tip: even if you prefer coding in JupyterLab, I recommend taking advantage of the debugger! The UI is actually very similar to PyCharm, and it's quite powerful.

What is your go to tool for debugging? Any favorites or personal insider tips people should know about? https://t.co/MrI9dJHiEW pic.twitter.com/uCU9Jp40G5

— Sebastian Raschka (@rasbt) June 14, 2022

Chalk: a python diagram library.

Docs: https://t.co/KMzU1wrujT
Git: https://t.co/rbhOh1M0OY

(a non-ml summer project with @DanOneata) pic.twitter.com/puCf5cTGFu

— Sasha Rush @ ICML (@srush_nlp) June 14, 2022

Perhaps it's true that "kindness and honesty is not the way of our modern workforce".

But perhaps if we can live our lives as if that's not the case, we can bring a little change into the world.

— Jeremy Howard (@jeremyphoward) June 13, 2022

we should all worry about this graph.

academia has its problems but I have never seen anything like the current AI/ML monoculture.

the movement from a peer-reviewed academic culture to a corporate, resource-intensive, hype-driven culture is likely partly responsible. https://t.co/1go6YxAQKV

— Gary Marcus 🇺🇦 (@GaryMarcus) June 13, 2022

The Missing Link: Finding label relations across datasets
abs: https://t.co/9C36CkXc3W pic.twitter.com/Tu3uQrHTH5

— AK (@_akhaliq) June 11, 2022

A new version of the {gt} package has been released! 🎉
Version 0.6.0 has even more features for creating and presenting summary tables in #rstatshttps://t.co/p2qEepdXrb

— RStudio (@rstudio) June 10, 2022

Cyclic TV Reference Paradox Finder. Only look at this website if you have time to spare. Which TV shows refer to which TV shows. So far, so easy to understand. Which TV shows refer to each other in a cyclical manner so that one can't be true? Source: https://t.co/yX0r0XCtxv pic.twitter.com/tY1UnAflRL

— Simon Kuestenmacher (@simongerman600) June 10, 2022