Deep Learning Monitor

Researchers in speech recognition, computer vision, and natural language processing in the 2000s were obsessed with… https://t.co/SYHHVFIAyv

2022-05-14 Super Hot

@DomSchiener

The scientific validation of #IOTA 2.0 is now published in three research papers explaining most of the innovations… https://t.co/4RwZGpAnz6

2022-05-05 Super Hot

Verdict from @icmlconf: 3 out of 3 ..... rejected. If I go by tweet statistics, ICML has rejeted every single paper this year 🤣

2022-05-16 Super Hot

Game over. Scale is essential to AI. https://t.co/naz31QVFRx

2022-05-09 Super Hot

@iota

[#Research] New #Milestone paper describing #consensus in #IOTA 2.0. "Tangle 2.0 Leaderless Nakamoto Consensus on t… https://t.co/KX4fQjZK98

2022-05-05 Super Hot

Actually, AI R&D has *3* flavors: 1. Applied AI: developing real-world product. 2. Press Release AI: maximally-impr… https://t.co/u3k6Ya0QHI

2022-05-14 Super Hot

@qikipedia

There are at least 23 academic papers that intentionally rickroll the reader. (Paper: https://t.co/pPFDaF2Yr7)

2022-05-10 Super Hot

Exactly. The idea that, somehow, people who develop DL methods aren't aware of their limitations, aren't trying to… https://t.co/b7qGMLiLzT

2022-05-08 Super Hot

@JeffDean

The new @googlemaps Immersive View feature is going to be pretty amazing (and uses Neural Radiance Fields, or NeRFs… https://t.co/8i1CzB0ooV

2022-05-12 Super Hot

Investments in DL *followed* empirical success. Let's not forget the 15 years between 1995 and 2010 during which t… https://t.co/d1OMoaJQjU

2022-05-04 Hot

Someone’s opinion article. My opinion: It’s all about scale now! The Game is Over! It’s about making these models b… https://t.co/vwBbAkQ33f

2022-05-14 Hot

OPT-175b: Open Pre-Trained language model with 175 billion paramaters is now available to the research community.… https://t.co/e4kHmP6WR0

2022-05-03 Hot

Free 449-page PDF eBook >> Understanding #MachineLearning — From Theory to #Algorithms: https://t.co/Nm19wd1X5n ———… https://t.co/v5t671qJpu

2022-05-04 Hot

Attention may be all you need. But short attention span is all you get. https://t.co/msZl1ymIIQ

2022-05-11 Hot

Openness is what makes progress in AI so fast. And accelerating. https://t.co/63tQvr3iE3

2022-05-09 Hot

@boredyannlecun @MetaAI As it turns out, you can't replace full-rank employees by low-dimensional embeddings. Embe… https://t.co/SWViWnecFn

2022-05-07 Hot

@ustx987 @icmlconf 1. Not every paper with my name on it is important. 2. I don't review often, but I'm a pretty ge… https://t.co/R2KYuiT1in

2022-05-16 Hot

@iota

[#Research] New paper: "Reality-based UTXO Ledger" 📚! This new UTXO Ledger model optimistically updates the ledger… https://t.co/Tqi9QjHk86

2022-05-04 Hot

@iota

[#Research] A new research paper is out 📚! Delve into a feedback approach to the design of an #AttackMitigation pol… https://t.co/jCI983HB4i

2022-05-04 Hot

@GoogleAI

Alpa is a framework that uses just one line of code to easily automate the complex model parallelism process for la… https://t.co/f5fq4dX6Ju

2022-05-04 Hot

@DeepMind

How do you tell an ML system to optimise an objective, but not by any means? E.g. optimise user engagement without… https://t.co/d6abiw4fq8

2022-05-06 Hot

Two years in the making by a talented, collaborative, and fun team, and with enormous help and support from many ot… https://t.co/MgaQIVjrLd

2022-05-12 Hot

As it turned out, solving the problem of learning hierarchical representations and complex functional dependencies… https://t.co/AA8VQfvmsa

2022-05-14 Hot

#infographic List of large Language Models for #NLProc + Learn more about such models: 1) https://t.co/XUHYA0goWG 2… https://t.co/TQ3gnxbiIl

2022-05-07 Hot

@GaryMarcus @wjb_003 @Tesla I dont engage in vacuous debates. Nor do I speculate about vague hypothetical proposals… https://t.co/ElK7Uy4Sr7

2022-05-08 Hot

@MaximZiatdinov @icmlconf To me, the benefit is pretty much zero. To the students and postdocs working with me who… https://t.co/b5R2kxMw96

2022-05-16 Hot

@DeepMind

3D priors have been utilised in DNNs via hardwired architectural designs. Research from #CVPR2022 shows how to inje… https://t.co/3SxtMYIA3k

2022-05-10 Hot

https://t.co/DszWHOIEo7 https://t.co/fYwEbo8xLG

Top 10 #Python Libraries for #DataScience in 2022: https://t.co/n81bqsAPDW by @simplilearn ———— #AI… https://t.co/vPQIRxy4ww

2022-05-08 Hot

@santisiri

2022-05-15 Hot

@GaryMarcus @Tesla According to independent studies in a few countries, Automatic Emergency Braking systems reduce… https://t.co/VFmMiKbzGY

2022-05-08 Hot

@chordbug

“Is this the simplest (and most surprising) sorting algorithm ever?” https://t.co/ucjbOCh0mA Abstract: We present… https://t.co/aokjApRS9p

2022-05-07 Hot

What it takes to successfully train a large language model, in excruciating details. There was open science. That… https://t.co/ImK8GMR3Lk

2022-05-04 Hot

But almost none of this work was concenred with the problem of learning representations. Features were assumed to… https://t.co/6KLE4A07WE

2022-05-14 Hot

When a humble Flamingo 🦩 meets a little Gato 🐈 https://t.co/sUT38itExr

2022-05-12 Hot

This led to a flurry of work on probabilstic generative models such as Hidden Markov Models in speech, Markov rando… https://t.co/zirAchsqlf

2022-05-14 Hot

Regularization in #MachineLearning: https://t.co/6KvYQbnjYZ ——————— #AI #DeepLearning #DataScience #BigData… https://t.co/2dfpAp0FGp

2022-05-10 Hot

Use *already solved* #DataScience projects for your real-world business problems now. Explore many examples here:… https://t.co/CuumIdU8Kq

2022-05-05 Hot

@GaryMarcus @Tesla You obviously never engineered nor built a working AI system. Arguments in favor "hybrid systems… https://t.co/zAbnlyQIqq

2022-05-08 Hot

Artificial Intelligence with #Python — a complete guide to building intelligent apps: https://t.co/ZQcEnw8NV4 —————… https://t.co/KlnXudncjD

2022-05-14 Hot

@GaryMarcus @Tesla Empty semantics. Obviously, *any* complete AI system, even one built around a big neural net, wi… https://t.co/Fbxv0mlTS3

2022-05-08 Hot

🦩explains my previous tweet. There are times when you feel your world has changed. As Scott Aaronson says, the dog… https://t.co/tqZfh12v55

2022-05-10 Hot

Man, does Twitter suck as a platform for explanatory pieces. For a more reasonable intellectual discussion platform… https://t.co/UPMbRJqzoK

2022-05-14 Hot

It's interesting to observe that almost none of this is relevant to today's top speech, vision, and NLP systems. 8/N

2022-05-14 Hot

Sobering trend. https://t.co/pFkaGL74RS

2022-05-08 Hot

Much of this was riding on previous work on Bayesian networks, factor graphs and other graphical models. That's how… https://t.co/XI39x3E7U4

2022-05-14 Hot

@boredyannlecun An efficient way to make bottleneck auto-encoders.

2022-05-13 Hot

@ISID_AI_team

少ないデータでディープラーニングを可能にする、「Few-shot learning」 (FSL)系の各種手法、その発展の歴史、応用事例、そして今後の期待について、分かりやすい解説と図解の論文でした A Comprehensi… https://t.co/VhhglKbOER

2022-05-16 Hot

Neural art by Meta AI artist-in-residence Sofia Crespo. In Time Square, NYC. Every night of May. Between 11:57pm an… https://t.co/yz4XRnxGPj

2022-05-07 Hot

There were debates at computer vision workshops about "generative models vs discriminative models". There were hero… https://t.co/K23krMdMFc

2022-05-14 Hot

🌟💯🚀Popular #MachineLearning Algorithms Explained With #Python Code: https://t.co/xc67ZU8pYt by @TheInsaneApp ————… https://t.co/zIJ6jZbPcE

2022-05-08 Hot

All one had to do was to compute some sort of log likelihood by linearly combining features, and then use one of th… https://t.co/9woiGPTTmP

2022-05-14 Hot

@pmddomingos No. But Gary is the one who insists on viewing this as "we vs them" conflict. I don't see a conflict.… https://t.co/oEFNMtCI7f

2022-05-09 Hot

This is not to say that accurate probabilistic inference is not useful. Comments? Opinions? 10/N

2022-05-14 Hot

🌟💯🚀Popular #MachineLearning Algorithms Explained With #Python Code: https://t.co/xc67ZU8pYt by @TheInsaneApp ————… https://t.co/eVusGrv3hV

2022-05-13 Hot

@intelnews

Intel’s 7nm Gaudi 2 processor delivers a major leap in performance of the highest end #DeepLearning training. Lea… https://t.co/U5kxGSEdAY

2022-05-10 Hot

In fact, "exponential family" pretty much means "shallow": the log-likelihood can be expressed as a linearly parame… https://t.co/vQ4FOcpp6a

2022-05-14 Hot

@DeepMind

In case you missed it...Flamingo 🦩 a new SOTA visual language model. Read more below ⬇️ Paper:… https://t.co/uG5xkkFqZ1

2022-05-09 Hot

Solving these scaling challenges is what will deliver AGI. Research focused on these problems, eg S4 for greater me… https://t.co/S4Ayil7LtP

2022-05-14 Hot

@WikiResearch

"OpenAlex: A fully-open index of scholarly works, authors, venues, institutions, and concepts", based on @Wikidata… https://t.co/nL49pqmuk0

2022-05-11 Hot

@ykilcher Large Language Model sometimes miss jokes, figures of speech, and subtelties in language. But so do many… https://t.co/dOleY0TEJz

2022-05-08 Hot

@dan_biderman I think factor graphs as a conceptual framework will become i creasingly relevant as deep learning sy… https://t.co/3WLrXyfKK7

2022-05-14 Hot

That was 11/N, N=11.

2022-05-14

If it was about babies,the US would have all the social programs listed here. Which *every* *single* liberal democr… https://t.co/8GtxwDUjAz

2022-05-04

MIT Researchers Create Tool to Forecast Multivariate Time Series: https://t.co/6qMT22I1C8 + Research paper:… https://t.co/9kmDhxsvXG

2022-05-15

@joisino_

主著論文 "Re-evaluating Word Mover's Distance" が ICML 2022 に Long Presentation として採択されました 🎉 正しく正規化すると単純な bag-of-words… https://t.co/xXWnOBX0qd

2022-05-15

Awesome *UPDATED* 309-page eBook >> "Patterns, predictions, and actions: A story about #MachineLearning"… https://t.co/EKTM2JakYx

2022-05-09

ML execution should not be thought of as a pipeline. #MachineLearning Execution is a Directed Acyclic Graph (DAG, l… https://t.co/0iTyKUPVBG

2022-05-04

Artificial Intelligence with #Python — a complete guide to building intelligent apps: https://t.co/ZQcEnw8NV4 —————… https://t.co/sNqRBNU3VM

2022-05-10

@yoavgo OPT-175A: "I am your father." OPT-175B: "noooooooooooo!"

2022-05-05

Get FREE Demo of 180+ end-to-end project solutions that you can use in your organization’s #MachineLearning and… https://t.co/aHG6YrUqlx

2022-05-05

#Python Examples of Popular #MachineLearning #Algorithms with Interactive #Jupyter Demos, plus the Background Math… https://t.co/vodul7iyj0

2022-05-13

This was going to be my first post-Covid in-person talk at a major international event. Sadly, I got Covid myself l… https://t.co/NzM3Wj2J7K

2022-05-10

The Most Important Fundamentals of #PyTorch to Know for #DeepLearning: https://t.co/qzkHtpJqB1 ————— #BigData… https://t.co/PWtKNLndcl

2022-05-10

One side builds things and tries to make them work. The other side has no choice but employ the strategy described… https://t.co/cpPHKDrHOY

2022-05-08

@GaryMarcus @wjb_003 @Tesla You actually had all the resources in the world, at some point. At least all the resour… https://t.co/mm8qneIns5

2022-05-08

Get these 7 FREE PDF eBooks at @DataScienceCtrl https://t.co/VWdHNT2IQw #abdsc ————— #BigData #Analytics… https://t.co/EIaqewL4tl

2022-05-05

An honor and privilege to be giving the Ernst Presidential Lecture tomorrow at the International Society of Magneti… https://t.co/zTDDARz2BD

2022-05-10

100+ Free #MachineLearning PDF Downloads — the Best #AI and #ML eBooks — compiled by @TheInsaneApp ————— #BigData… https://t.co/kCBFtnxcFH

2022-05-16