Just made a user-friendly Python library for using our CORAL (COnsistent RAnk LOgits) in @PyTorch. Includes examples showing how to equip deep neural networks (convolutional neural networks and multilayer perceptrons) for ordinal regression: https://t.co/JUqZ7ZYRHN pic.twitter.com/Qkvq18uZ0W
— Sebastian Raschka (@rasbt) November 16, 2020
This chart by @gzeromedia shows how ridiculous the accusation of voter fraud in the US election of 2020 is. But facts aren't too important apparently... pic.twitter.com/IDjfxGZLkG
— Simon Kuestenmacher (@simongerman600) November 13, 2020
We've started a great collaboration between @PyTorch and @MLflow, to bring a rich set #MLOps functionality to PyTorch users. We've been working on this with the PyTorch team for a while and we're super excited to release a first wave of integrations today: https://t.co/UnZ6fSOec3
— Matei Zaharia (@matei_zaharia) November 12, 2020
PyTorch is excited to partner with OpenMined to make private AI education more accessible for everyone. https://t.co/6fIHaEhzaM
— PyTorch (@PyTorch) November 12, 2020
Aspiring academics, look at politics now to see the exact opposite of everything we value: arguing to win (vs to learn), inoculation against opposing evidence (vs seeking it out), severing ties with adversaries (vs sharing drinks). Despite flaws, research communities are special.
— Zachary Lipton (@zacharylipton) November 11, 2020
A simple prescription that will improve your models: When using LayerNorm, do mean subtraction *before* rather than after the affine transformation.
— Jascha Sohl-Dickstein (@jaschasd) November 11, 2020
This, and an in-depth empirical investigation of statistical properties of common normalizers in https://t.co/poAWNmpSah pic.twitter.com/Ojcj0hsnTM
This is the first time I've seen geographic data represented in this way. While COVID deaths in the northern hemisphere are rising they are declining in the southern hemisphere. Took me a second to read the chart correctly. Source: https://t.co/RnJO5fUhJa pic.twitter.com/VGSUgRmxxJ
— Simon Kuestenmacher (@simongerman600) November 11, 2020
Understanding the Difficulty of Training Transformers (https://t.co/lN5JccNOlz, @LiyuanLucas)
— Sasha Rush (@srush_nlp) November 11, 2020
Studies instability of transformer. Dives into the dark arts of NN stability, impact of layer norm / residuals. Isolates residual paths as a main cause. Trains 60 layers transformers. pic.twitter.com/SChDwCo4xa
"Deep-learning-in-cloud".Just stumbled about this really useful resource. This list compiles a list of 41 different services for using GPUs for DL! Especially useful for students looking to get into using GPUs for their class projects: https://t.co/VI1JnQ9Z9O pic.twitter.com/xNyNGBrjmg
— Sebastian Raschka (@rasbt) November 11, 2020
Ever wondered how much stay-at-home slows down COVID-19? In our latest @nature paper we use smartphone data to model mobility of 98 million people and detect hotspots, track COVID-19, and guide reopening strategies. Try the model at https://t.co/1VgPKwViZt pic.twitter.com/0sbbGZo8ls
— Jure Leskovec (@jure) November 10, 2020
These text-to-image results look pretty cool.https://t.co/CPkC56bxst https://t.co/TbAaYZ9pwl pic.twitter.com/GNShK9NSmp
— hardmaru (@hardmaru) November 10, 2020
Wave-Tacotron: Spectrogram-free end-to-end text-to-speech synthesis
— AK (@ak92501) November 10, 2020
pdf: https://t.co/gx4xC0fWrz
abs: https://t.co/zWz2ZIPVOR
project page: https://t.co/CEexNbhUs8 pic.twitter.com/30AhWoVsDo
