Machine learning research @GoogleAI, Opinions mine.
Joined July 2018
- Tweets 27
- Following 102
- Followers 611
- Likes 48
1 Photos and videos
André Susano Pinto retweeted
🚀Gemma4 12B🚀
We made it great by training a simpler model.
No vision or audio encoders. Easier said than done.
Running exploratory experiments to a final model is always interesting. Joint work with @mtschannen @AndreasPSteiner @confusezius @kmisiunas and the whole Gemma team.
1
5
211
André Susano Pinto retweeted
20 Dec 2024
Check out our detailed report about *Jet* 🌊 - a simple, transformer-based normalizing flow architecture without bells and whistles.
Jet is an important part of JetFormer's engine ⚙️ As a standalone model it is very tame and behaves predictably (e.g. when scaling it up).
20 Dec 2024
With some delay, JetFormer's *prequel* paper is finally out on arXiv: a radically simple ViT-based normalizing flow (NF) model that achieves SOTA results in its class.
Jet is one of the key components of JetFormer, deserving a standalone report. Let's unpack: 🧵⬇️
8
32
4,114
20 Dec 2024
Jet, the tool in JetFormer. A coupling normalizing flow where the blocks are powered by ViT.
Simple, scalable and it works!
20 Dec 2024
With some delay, JetFormer's *prequel* paper is finally out on arXiv: a radically simple ViT-based normalizing flow (NF) model that achieves SOTA results in its class.
Jet is one of the key components of JetFormer, deserving a standalone report. Let's unpack: 🧵⬇️
1
3
6
2,067
20 Dec 2024
Making new simple things requires attention to detail. From numeric precision and unexpected bugs deep in the stack. But now there is a precedent which includes paper, numbers and code.
Hope it helps people go hammer some nails🔨
2
189
André Susano Pinto retweeted
5 Dec 2024
Welcome PaliGemma 2! 🤗
Google released PaliGemma 2, best vision language model family that comes in various sizes: 3B, 10B, 28B, based on Gemma 2 and SigLIP, comes with transformers support day-0 🎁
Saying this model is amazing would be an understatement, keep reading ✨
28
249
1,660
167,209
André Susano Pinto retweeted
5 Dec 2024
🚀🚀PaliGemma 2 is our updated and improved PaliGemma release using the Gemma 2 models and providing new pre-trained checkpoints for the full cross product of {224px,448px,896px} resolutions and {3B,10B,28B} model sizes.
1/7
4
52
261
61,928
2 Dec 2024
Did you try to get an auto-regressive transformer to operate in a continuous latent space which is not fixed ahead of time but learned end to end from scratch?
Enter JetFormer: arxiv.org/abs/2411.19722 -- joint work in a dream team: @mtschannen and @__kolesnikov__
2 Dec 2024
Have you ever wondered how to train an autoregressive generative transformer on text and raw pixels, without a pretrained visual tokenizer (e.g. VQ-VAE)?
We have been pondering this during summer and developed a new model: JetFormer 🌊🤖
arxiv.org/abs/2411.19722
A thread 👇
1/
1
3
37
4,828
21 Nov 2019
Feels great to start adding diversity to the available pre-trained visual representations. Especially when it has considerable impact for problems with a smaller number of examples available or hard to collect.
21 Nov 2019
We've looked into representation learning for #RemoteSensing with different datasets and fine-tuning using in-domain data. See paper with datasets and models included 🔋: arxiv.org/abs/1911.06721 with @ASusanoPinto, @XiaohuaZhai and @neilhoulsby.
4
We’re pleased to release the Visual Task Adaptation Benchmark (VTAB), a diverse, realistic, and challenging protocol to measure progress towards universal visual representations. Learn all about it below. goo.gle/2Noutb9
3
113
327
11 Mar 2019
#TensorFlowHub helping fast experimentation and making ML models that go to space.
Amazing article showing how accessible #DeepLearning is becoming. Model trained with transfer learning and "#TensorFlow For Poets" codelab #tfhub. Converted to #TFLite and now deployed on International
Space Station🚀 - TensorFlow Lite is Going to Space - medium.com/tensorflow/tensor…
1
2
André Susano Pinto retweeted
6 Mar 2019
BigGAN-deep pretrained models are now publicly available for download on TFHub!
tfhub.dev/s?q=biggan
22 Feb 2019
BigGAN has been accepted for oral presentation at ICLR2019! We've uploaded a revision of the paper with an improved architecture, BigGAN-deep: 4x the depth, 50% *fewer* parameters, and even better performance.
openreview.net/pdf?id=B1xsqj…
1
45
187
André Susano Pinto retweeted
5 Mar 2019
A new, multilingual version of the Universal Sentence Encoder (USE) model is now available on #TFHub!
Check it out here → bit.ly/2J7ZJuX
55
197
André Susano Pinto retweeted
12 Nov 2018
The BigGAN generators from our paper arxiv.org/abs/1809.11096 are now available on TF Hub (tfhub.dev/s?q=biggan). Try the Colab demo at: colab.research.google.com/gi…
13
488
1,369
André Susano Pinto retweeted
5 Oct 2018
Enjoying the Workshop by Google engineer Elizabeth Kemp: Transfer Learning with TensorFlow Hub 👩💻
#womencourage18 #ML #TensorFlow
4
8
17 Sep 2018
Our team hopes the new frontend helps more people find and use cutting-edge research modules :) #TensorFlowHub #transferlearning
17 Sep 2018
We are launching a new web experience for TensorFlow Hub! Check out tfhub.dev and explore our modules, including some new additions like the FasterRCNN for object detection.
Learn more on the post ↓ medium.com/tensorflow/a-new-…
4
6 Sep 2018
Great to have image embedding modules trained on datasets other than just ImageNet.
5 Sep 2018
Winners of the @inaturalist Challenge 2017 released their model on #TensorflowHub showcasing advantages of transfer learning! #tfhub #transferlearning
Check it out here ↓ bit.ly/2NRVAsM
1
André Susano Pinto retweeted
20 Aug 2018
Had 92% accuracy for predicting salary range with job description text using #tensorflowhub . #MachineLearning #TransferLearning #AI . Thank you #tfhub team for making this transfer learning module available.
1
2
André Susano Pinto retweeted
14 Aug 2018
TensorFlow v2.0 is coming - with a focus on ease of use!
groups.google.com/a/tensorfl…
20
325
990
André Susano Pinto retweeted
15 Aug 2018
Oh hey, all the code you need to instantiate a @TensorFlow Hub module fits in a tweet!
embeddings = hub.text_embedding_column(
"descriptions",
module_spec="tfhub.dev/google/universal-s…"
)
Details in this blog post: medium.com/tensorflow/buildi…
2
13
33
André Susano Pinto retweeted
10 Aug 2018
Yoshua Bengio: "You need space in your week to think, to not work on programming or writing or even reading. Just think about the big questions that bug you." cifar.ca/news/news/2018/08/0…
6
127
535