Deep Learning Monitor

Reinforcement Learning for Improving Agent Design: What happens when we let an agent learn a better body design tog… https://t.co/WEpALasSVv

2018-10-10 Super Hot

@DeepMindAI

We've just released our Graph Nets library for deep learning on graphs: https://t.co/BCaErooanD. For more about Gra… https://t.co/abfTApr41h

2018-10-19 Super Hot

@lmthang

A new era of NLP has just begun a few days ago: large pretraining models (Transformer 24 layers, 1024 dim, 16 heads… https://t.co/7TQnjIDVV1

2018-10-12 Super Hot

If we remove all design constraints, the optimizer came up with a really tall bipedal walker robot that “solves” th… https://t.co/Wus3YEpeOM

2018-10-11 Super Hot

So many papers applying deep learning to theoretical and experimental physics! Fascinating. https://t.co/F8YfsFQaDL

2018-10-08 Super Hot

Large Collection of Cheat Sheets for #DataScience and #MachingLearning, including #Python, R, #NeuralNetworks, Nump… https://t.co/NQJNOpvSUj

2018-10-14 Super Hot

Holodeck is a high-fidelity simulator for reinforcement learning built on top of Unreal Engine 4, with an OpenAI-Gy… https://t.co/iaZ1a42IwL

2018-10-08 Super Hot

The essence of strategy is choosing what not to do.

2018-10-13 Super Hot

Machine Learning cheatsheet for Stanford's CS 229, in various languages! original: https://t.co/U8jy3psnuB translat… https://t.co/gqTdIvrkvA

2018-10-17 Super Hot

@kaggle

Bookmark: This 218-page Intro to Probabilistic Programming. 🤓https://t.co/QVybd8us4A Many thanks to 🙏 @hyang144, J… https://t.co/xPdudEMXcf

2018-10-18 Super Hot

PDF version of brilliant #MachineLearning cheatsheets covering the content of Stanford's CS 229 class::… https://t.co/jeDAR55NKF

2018-10-17 Super Hot

Not following the API https://t.co/t9dVWAqyoI

2018-10-20 Hot

The discriminator often knows something about the data distribution that the generator didn't manage to capture. By… https://t.co/1eyne8axqy

2018-10-17 Hot

@java

Free Standford Machine Learning Lectures Lecture collection on deep learning architectures with a focus on learni… https://t.co/t0HiBDQ29U

2018-10-07 Hot

PyTorch, now available on Google Cloud. TPU support coming up. https://t.co/gZayUce67q

2018-10-17 Hot

@zacharylipton

Hey everyone. Stoked to report that we've blown away recent NLP benchmarks with a new sentence embedding: "Efficien… https://t.co/NW2LTIxrkv

2018-10-17 Hot

@AndrewYNg

Every time there's a technological revolution like AI, it gives us a chance to remake large parts of the world. Spo… https://t.co/CEFhmbxQxS

2018-10-08 Hot

@AndrewYNg

After the successful launch in Frisco TX earlier this year, https://t.co/3tAfFJgG0d is now launching a new commerci… https://t.co/fqJChTxJpL

2018-10-19 Hot

30 Free Courses in #NeuralNetworks, #MachineLearning, #Algorithms, and #AI : https://t.co/LDbHoNxec4 #abdsc… https://t.co/eG3E1FQOZr

2018-10-12 Hot

@demishassabis

Had a really enjoyable conversation at @royalacademy speaking on a new topic: creativity and AI (new stuff from 29m… https://t.co/x6QMrTkIkt

2018-10-14 Hot

Trellis Networks for Sequence Modeling: They propose a bridge between RNN & CNN, a temporal CNN with special struct… https://t.co/63pv0KkjxD

2018-10-17 Hot

@nntaleb

Philosophy of Probability: How we can link the speed of convergence of the law of large numbers to the problem of i… https://t.co/rVj28ol7Ua

2018-10-16 Hot

@DeepMindAI

We should be optimistic but never complacent about the positive potential of AI, DeepMind COO @lilaibrahim writes i… https://t.co/Nyd01kofp0

2018-10-19 Hot

Some people are doing the exact reverse. I’ve seen the term “linear regression” replaced by “an AI Equation”! https://t.co/wUXoIO8MJL

2018-10-14 Hot

“The essential key for writing is to write regularly. Like it or not, great ideas come often by writing. Waiting fo… https://t.co/YyRKFzfdQf

2018-10-16 Hot

Data Science Mindmap: https://t.co/OoWsx87x7S #abdsc #BigData #Analytics #DataScience #MachineLearning… https://t.co/61UgzkVFzx

2018-10-15 Hot

@kchonyc

the paper behind BERT is now online: https://t.co/kKDzq3GF5W BERT: Pre-training of Deep Bidirectional Transformer… https://t.co/AkGkqXklHx

2018-10-12 Hot

At one point in time, nature only cared about solving vision problems. https://t.co/Xi0IYYAMfp

2018-10-14 Hot

How to build your own #NeuralNetworks from scratch in #Python — Beginner’s guide to understanding the inner working… https://t.co/khTzCw6G8a

2018-10-08 Hot

The #DataScientists short guide to Artificial #NeuralNetworks: Part 1: https://t.co/33HeDYVnX4 Part 2:… https://t.co/ncL2bM09gx

2018-10-07 Hot

Un documentaire sur l'IA diffusé sur Arte hier soir, où je suis filmé sous tous les angles dans mon environnement n… https://t.co/BKnpn068cC

2018-10-07 Hot

Metaphor for well-funded Silicon Valley startups: “Growth at all costs, land at an exit opportunity!” https://t.co/iRq2OLbgu6

2018-10-12 Hot

We are at a point where car owners are excited about neural net upgrades and software maintenance.… https://t.co/RuspuibAHP

2018-10-15 Hot

@rbhar90 The funny thing is, from my point of view, figuring out how to use DL as a building block was what we did… https://t.co/SNY7L7gtys

2018-10-17 Hot

@boredyannlecun Well, actually, this is more right than you think: A multiplication *is* a convolution of one mult… https://t.co/GexWLz1aiO

2018-10-20 Hot

“Recycle-GAN”: Unsupervised Video Retargeting. Translation from John Oliver to Stephen Colbert, and a synthesized f… https://t.co/Bttepx7ktx

2018-10-12 Hot

Good read We'd Like To See #DeepLearning Hand To Hand With Deep Feeling https://t.co/rbRoRTLYbN #fintech… https://t.co/SPHQLT70bD

2018-10-14 Hot

GPU-Accelerated Robotic Simulation for Distributed Reinforcement Learning: “With one GPU and CPU core, we are able… https://t.co/PSCtTlAFg2

2018-10-16 Hot

NEAT in PyTorch! Makes it easier to build GPU-accelerated versions of CPPN-NEAT for generating psychedelic neural a… https://t.co/zMRDw1CcBi

2018-10-11 Hot

@NandoDF

The neural networks for deep RL just got a lot bigger. Evidence that RL scales to train nets that must be massive t… https://t.co/UdILVd4O5x

2018-10-17 Hot

Human-Augmented Reality https://t.co/KAxQ6dqBMF

2018-10-16 Hot

Boston Dynamics’ #Atlas #robot can now do #parkour, don’t bother running away https://t.co/MGYaCnE8XT @mmmmurf… https://t.co/sM4mVzCeHX

2018-10-12 Hot

Test set’s data distribution vs Real world’s data distribution https://t.co/jYE1ULeids

2018-10-12 Hot

Okay this freaked me out 👀 https://t.co/A2iue6qTHY

2018-10-09 Hot

Like space monkeys ready to be shot into orbit, PhD students are the backbone of modern scientific and technologica… https://t.co/UApXSVIFXl

2018-10-13 Hot

PSA. If you’re a US green card holder, requirements to travel to Canada have changed recently. I imagine many of my… https://t.co/sXcUfnltbY

2018-10-16 Hot

@Rainmaker1973

A domino can knock over another domino about 1.5x larger than itself. Here are the conditions under which a row of… https://t.co/fNCXnliqyj

2018-10-07 Hot

@yoavgo

language modeling is not unsupervised learning.

2018-10-12 Hot

CINIC-10 is not ImageNet or CIFAR-10: Well-motivated design of a drop-in replacement for CIFAR-10!… https://t.co/Y6ubQbLfiD

2018-10-09 Hot

An R lovers paradise! >> Download the most fabulous R Cheat Sheets from @Rstudio: https://t.co/eY55zVyXI7 #abdsc… https://t.co/6gZb1d877d

2018-10-19 Hot

Train RL agents to play HTML5 games using Gym interface! https://t.co/QB5CQLl3Ek https://t.co/MF5FFSYCt3

2018-10-19 Hot

@NandoDF

So good https://t.co/liUq3pAdv9

2018-10-20 Hot

@ajitjohnson_n

Far beyond Pluto, astronomers have found a new dwarf planet that follows the most distant orbit yet confirmed. This… https://t.co/30E0iU0EMn

2018-10-09 Hot

Both @catherineols and I set our Google employee profile pictures to the cabbage head sample for a few days https://t.co/wZj2cWJXzj

2018-10-17 Hot

Inflation in SF vs Tokyo https://t.co/GOhyBsuRYa

2018-10-07 Hot

@RichardSocher

Flying close to the beautiful #clouds above the gorgeous coast of Central #California #paramotor https://t.co/g6EbeD89R5

2018-10-15 Hot

Apparently there were over 7000 papers submitted to AAAI 2019 conference, currently going through the peer review p… https://t.co/lToWFWfdB0

2018-10-13 Hot

@skdh

New paper estimates 10^723 standard models can be constructed in string theory. The zeroes on that number, fully ty… https://t.co/r6RCOX4QvB

2018-10-20 Hot

@nvidia

The deep-learning revolution sweeping the globe started with processors - GPUs - originally made for gaming. With o… https://t.co/QaLQr5mPlK

2018-10-15 Hot

@demishassabis

The study made the Editor's Choice in Science, nice short summary in the neuroscience section:… https://t.co/mncow7OQMw

2018-10-12 Hot

The #Robots Are Coming— To Clean Up Your Disgusting Room https://t.co/1reSxask6W #fintech #insurtech #AI… https://t.co/UYjK4k0ofY

2018-10-18 Hot

ICML2019 is following NIPS’ footsteps to “reward” good reviewers: “Conference tickets will be reserved for a limit… https://t.co/xo2FxYZPig

2018-10-17 Hot

There must be tens of thousands of PhD students working on AI/ML research problems. A large portion of them will be… https://t.co/ajqgaGLhyo

2018-10-13 Hot

Generative Tree in a Tweet 🌳 https://t.co/AxSBt2mcgt

2018-10-17

Another good guide for newcomers: “How to Write a Technical Paper” (PDF) https://t.co/P6L39ZddtD

2018-10-16

I think this agent got scooped! 🤖🤣 (taken from https://t.co/CBEimebVHv https://t.co/EeGNJIQ7pr) https://t.co/z1Kgo5ZIGI

2018-10-11

@MIT_CSAIL

DeepMind just released their Graph Nets library for deep learning on graphs: https://t.co/8gSKFW3TOZ Related paper:… https://t.co/kNFO0x338a

2018-10-19

★★★This Google Tutorial (96 slides total) on #MachineLearning is the best: https://t.co/pPvmPHGiFj by @jason_mayes… https://t.co/u6VKcqEEUl

2018-10-10

How #Amazon is using #AI in almost everything it does https://t.co/PsoR6gFVrJ #fintech #insurtech… https://t.co/0RXMaJakUz

2018-10-08

When you train SpotMinis using reinforcement learning inside a simulation environment ... https://t.co/ls2mIqlfvH

2018-10-15

Artificial Intelligence: Mankind's Last #Invention https://t.co/pTon5aKPM7 #fintech #insurtech #AI… https://t.co/RUT0cNsbER

2018-10-19

Good one The Rise of #AI https://t.co/tFWAwfeCJm #fintech #insurtech #ArtificialIntelligence #MachineLearning… https://t.co/u9TgKyKbwy

2018-10-11

@NandoDF

@karpathy Frowned upon?! We’d be happy to have you here starting at 2pm 😂

2018-10-19

"La plus belle histoire de l'intelligence" A book I co-authored with Stanislas Dehaene and Jacques Girardon is com… https://t.co/pxBsS9XAQN

2018-10-16

Art-Driven Advertising https://t.co/9jaT2jyIMw https://t.co/L76v17XcMO

2018-10-12

SomeNET, MyGAN, XYZ is All You Need, Learning to learn by swimming to swim etc., is ＊SO＊ 2017, and out of fashion.… https://t.co/ATYoT66o1m

2018-10-19

When rewarded for small leg size, the agent learned a lean minimal design where every inch matters. It also learned… https://t.co/Rac4f7PkZy

2018-10-11