Abhirup Das @iam_abhirup

Joined October 2019

Tweets

60
Followers

15
Following

118
Likes

375

Ben Tossell @bentossell

3 years ago

All the best examples of ChatGPT, from OpenAI:

373 3K 14K 0 8K

Andrej Karpathy @karpathy

5 years ago

Taming Transformers for High-Resolution Image Synthesis github.com/CompVis/taming… impressive work/results! (also fun to see a shoutout and my minGPT code used for the transformer :))

12 215 1K 0 247

Download Image

Today we announce a novel, open-source method for text generation tasks (e.g., summarization or sentence fusion), which uses edit operations instead of generating text from scratch, leading to less errors and faster model execution. Read about it below. goo.gle/38XfRXU

9 280 825 0 124

Andrej Karpathy @karpathy

7 years ago

"An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling" arxiv.org/abs/1803.01271 good to see more papers on the topic - always start with a CNN before reaching for an RNN. You'll be surprised with how far you can get.

9 333 1K 0 95

David Berthelot @D_Berthelot_ML

6 years ago

FixMatch: focusing on simplicity for semi-supervised learning and improving state of the art (CIFAR 94.9% with 250 labels, 88.6% with 40). arxiv.org/abs/2001.07685 Collaboration with Kihyuk Sohn, @chunliang_tw @ZizhaoZhang Nicholas Carlini @ekindogus @Han_Zhang_ @colinraffel

5 234 857 0 210

Download Image

hardmaru @hardmaru

6 years ago

Ants use the rate at which they meet and smell other ants to decide what to do next. A neuron uses the rate at which it is stimulated by other neurons to decide whether to fire. Memory arises from changes in how they connect and stimulate each other. 🐜🧠 aeon.co/ideas/an-ant-c…

3 57 317 0 25

Neil Lawrence @lawrennd

6 years ago

1/ This shows how far the field has regressed in its understanding of probability. It's not a controversial opion, it's the opinion of someone who hasn't understood that a prior over weights in a neural network induces a prior over functions.

Carles Gelada (at ICML) @carlesgelada

6 years ago

27 101 583 0 199

4 34 224 0 53

Abhirup Das @iam_abhirup

6 years ago

technologyreview.com/s/615054/deepm…

0 0 0 0 0

Kevin Musgrave @KevMusgrave

6 years ago

PyTorch Metric Learning now available on Anaconda! Installation: "conda install pytorch-metric-learning -c metric-learning" View more details here: github.com/KevinMusgrave/… #DeepLearning #MachineLearning #PyTorch #Python #ComputerVision #conda

1 197 520 0 87

Download Image

Graham Neubig @gneubig

6 years ago

Happy to release NN4NLP-concepts! github.com/neulab/nn4nlp-… It's a typology of important concepts that you should know to implement SOTA NLP models using neural nets: github.com/neulab/nn4nlp-… 1/3 We'll reference this in CMU CS11-747 this year, trying to maximize coverage. 1/3

Graham Neubig @gneubig

6 years ago

8 71 293 0 51

Download Image

2 108 314 0 74

Download Image

elvis @omarsar0

6 years ago

📰 NLP Newsletter (Issue #1): Tokenizers, TensorFlow 2.1, TextVectorization, TorchIO, NLP Shortfalls,…📰 featuring: @seb_ruder, @viglovikov, @huggingface, @TensorFlow, @zacharylipton, @timnitGebru, @gradientpub, GitHub: github.com/dair-ai/nlp_ne… Medium: medium.com/dair-ai/nlp-ne…

3 109 391 0 63

Jenn Wortman Vaughan @jennwvaughan

6 years ago

Ok, people! Are you looking for something to read at the intersection of machine learning and HCI? Three new papers posted online today that you should check out, all with my amazing colleague/BFF @hannawallach! Ready? I'm gonna try a thread!

5 71 330 0 84

DeepLearning.AI @DeepLearningAI

6 years ago

Recent language models like BERT and ERNIE rely on trendy layers based on transformer networks and require lots of compute. Deep learning researcher @Smerity shows these layers and enormous GPUs may not be necessary: bit.ly/2QZXxWV

5 96 366 0 59

Download Image

Brian Roemmele @BrianRoemmele

6 years ago

Consider: millions of years ago our antecedents gave a massive sacrifice of their left hemisphere. We lost a tremendous amount of short term memory and replaced it with Broca’s, Wernicke & the phonological loop. But why? So we can—talk. Thus chimpanzees can do this—we can’t:

836 30K 87K 0 7K

Download Video

Chip Huyen @chipro

6 years ago

These books shaped my last decade, and since it was my coming-of-age, these books shaped who I am. I've coerced some of my friends to show me their lists and loved everything I saw. I would love to see your list too :-) huyenchip.com/2019/12/28/boo…

14 90 646 0 152

Sara Hooker @sarahookr

6 years ago

What does a pruned deep neural network "forget"? Very excited to share our recent work w Aaron Courville, Yann Dauphin and @DreFrome weightpruningdamage.github.io

8 158 590 0 130

Download Image

Yann LeCun @ylecun

6 years ago

There are now 3 papers that successfully use Self-Supervised Learning for visual feature learning: MoCo: arxiv.org/abs/1911.05722 PIRL: arxiv.org/abs/1912.01991 And this, below. All three use some form of Siamese net.

Aäron van den Oord @avdnoord

6 years ago

4 193 714 0 93

Download Image

2 327 999 0 204

Chip Huyen @chipro

6 years ago

To learn how to design machine learning systems, I find it really helpful to read case studies to see how great teams deal with different deployment requirements and constraints. Here are some of my favorite case studies.

23 514 2K 0 861

Open Data Science @_odsc

6 years ago

An artificial intelligence model developed at MIT shows in striking detail what makes some images stick in our minds. #DataScience #AI #ArtificialIntelligence hubs.ly/H0lQwyS0