Real Person to Anime
— Ian Goodfellow (@goodfellow_ian) June 7, 2018
[ethics] Google AI Principles
Today we’re sharing our AI principles and practices. How AI is developed and used will have a significant impact on society for many years to come. We feel a deep responsibility to get this right. https://t.co/TCatoYHN2m
— Sundar Pichai (@sundarpichai) June 7, 2018
In addition, we’re offering our technical practices to actually implement these principles - we hope everyone finds these informative and useful as we move forward as a community. #GoogleAI https://t.co/UwVI8CPxGj
— Jeff Dean (@JeffDean) June 7, 2018
(Bonus)
Should #AI researchers work on projects that help militaries? Check out @Gregory_C_Allen in @nature tackling this tough question: https://t.co/JLE9Mhsheb
— Michael C. Horowitz (@mchorowitz) June 7, 2018
Visualization
Shift in teen social media usage since 2015, based on @pewinternet surveys https://t.co/CRd9bCTjyc pic.twitter.com/UaGA9JSMMX
— Nathan Yau (@flowingdata) June 7, 2018
Small multiple flow in @tableaupublic showing how teams advanced throughout the NBA playoffs. Write up: https://t.co/SKs6AOMiIZ & viz: https://t.co/3tnoCjyPAD pic.twitter.com/G5mGJyBd05
— Chris DeMartini (@demartsc) June 5, 2018
Cool Sankey diagram of World Cup 2018 simulations. All that green is Germany. Model has Germany at ~25%, followed by Brazil (~12%), Spain (~11%), and Argentina (~10%). Source: https://t.co/wlWgseqtEX pic.twitter.com/NFbhkn2IDj
— Luke Bornn (@LukeBornn) June 7, 2018
An interesting case study that uses multiple charts to show different insights into "visualizing pedestrian data" in this article by @morphocode (starts about halfway down the page) https://t.co/BBo8EA4Lzk pic.twitter.com/O5Rz2WNcpW
— Nadieh Bremer (@NadiehBremer) June 7, 2018
Apology from Ferenc Huszár
(Long thread. Click on the tweet to view the full thread.)
Few days ago I tweeted things I should not have. It was bad, I regret and apologize.
— Ferenc Huszár (@fhuszar) June 7, 2018
This sort of stuff undermines the effort of colleagues, and my own, to articulate the important role various disciplines play in taking ML forward and to create a welcoming and healthy community.
Notable Research
New AI safety paper - "Measuring and avoiding side effects using relative reachability": https://t.co/OIfaY3xIib
— DeepMind (@DeepMindAI) June 7, 2018
New paper on single trial dimensionality reduction and demixing in #neuroscience through tensor decompositions now published in @NeuroCellPress Congrats to @ItsNeuronal for leading this and thanks to @shenoystanford and Schnitzer lab for collaborating! https://t.co/EAxNd7ARj2
— Surya Ganguli (@SuryaGanguli) June 8, 2018
This weekend I compared some #sentence #similarity methods. @prfsanjeevarora's Smooth Inverse Frequency with #word2vec came out as the top pick. https://t.co/bwhbrGpWSP #NLProc #AI #deeplearning #textmining pic.twitter.com/vRIi1sP7dC
— Yves Peirsman (@yvespeirsman) May 2, 2018
Just learned that our paper "Synthesizing Programs for Images using Reinforced Adversarial Learning" (aka SPIRAL) got a long talk at @icmlconf. If you haven't already, go check it out: https://t.co/1KA8brgIFg pic.twitter.com/T0LP8pGgwM
— Yaroslav Ganin (@yaroslav_ganin) June 7, 2018
Check out my work at @GoogleAI on a linear complexity tSNE implementation for TensorFlow.js on the web! https://t.co/uXLEin9T5o
— Nicola Pezzotti (@nicolapezzotti) June 7, 2018
It has been great to collaborate with so many amazing engineers and researchers @zzznah @tafsiri @nsthorat @dsmilkov!
Tutorials and Resources
ml5.js is a machine learning library for creative coders that is built on top of tensorflow.js, with an API that is heavily inspired by Processing and p5.js. The examples are very readable and fun. Here is a char-rnn demo trained on Ernest Hemingway: https://t.co/LFlAabAzlh pic.twitter.com/pY918migHe
— hardmaru (@hardmaru) June 8, 2018
Video of World of Science Festival panel discussion on AI.
— Yann LeCun (@ylecun) June 7, 2018
Moderated by Tim Urban, with panelists Susan Schneider, Peter Tse, Max Tegmark and yours truly.
There is a pretty good article with the main points of the debate at Adventures In Poor... https://t.co/SGcVVFJ97R
Command Line Tricks For Data Scientists https://t.co/DRCGCEYeeN pic.twitter.com/Stvws3jOxI
— KDnuggets (@kdnuggets) June 7, 2018
more bash tricks pic.twitter.com/HPeriSzIIi
— 🔎Julia Evans🔍 (@b0rk) June 7, 2018
💥 trove of resources!
— Mara Averick (@dataandme) June 7, 2018
📊 "Open Access VIS - collection of open access visualization papers, material, and data"https://t.co/OEPWhbb94h via @sharoz #dataviz #ieeevis
[see same for #eurovis via @jamesscottbrown https://t.co/e73HtXHS6m] pic.twitter.com/25kjTYYPvT
Announcing https://t.co/SAb4OLLRTB! We are building a repository of study guides targeting consequential papers. Check it out, learn something in-depth, and help us build the next one. @avitaloliver @suryabhupa @kumarkagrawal @cinjoncin
— Depth First Learning (@DepthFirstLearn) June 6, 2018
Python
"Optional Static Typing for #Python"
— Damian Gryski (@dgryski) June 7, 2018
By @gvanrossum https://t.co/6qPEbRdkHU
#rstats
#RStats — A nice free 📚 about Handling Strings with R : https://t.co/zXOJYLsmVE
— Colin Fay (@_ColinFay) June 7, 2018
Miscellaneous
1989: Self-driving car with 4 neurons
— hardmaru (@hardmaru) June 7, 2018
2018: Play Atari game with 6 neurons#progress https://t.co/Yu7aO9lV0B
How Facebook scales AI: a nice piece at ZDNet following a keynote at the International Symposium on Computer Architecture (ISCA) by Kim Hazelwood, head of Facebook's AI Infrastructure group.
— Yann LeCun (@ylecun) June 7, 2018
"Most of Facebook's two... https://t.co/qVEdQm4D8R
The lack of scientific thinking affects every subfield of CS. A recent paper about computer security points out what should have been obvious: we've been trying to theorem-prove our way to scientific knowledge, but we need experiments and falsifiablily. https://t.co/GkMU5SIpxy
— Arvind Narayanan (@random_walker) June 7, 2018
As a data science type person, if there was one thing I could say to most companies, it would be this: Store your data properly, goddammit.
— Emma Vitz (@EmmaVitz) June 7, 2018
We can talk about ML and AI all you like, but if the data's not there or the quality's not there, that's all for nothing.