Free Live Course: Deep Learning with PyTorch

https://www.freecodecamp.org/news/free-deep-learning-with-pytorch-live-course/

video: https://www.youtube.com/watch?v=vo_fUOk-IKk

🗣 @AI_Python_arXiv
✴️ @AI_Python_EN
❇️ @AI_Python

Introduction to Neural Networks in Python | Tensorflow/Keras

https://morioh.com/p/abae4875bdd2/

SymJAX: symbolic CPU/GPU/TPU programming.

docs: https://symjax.readthedocs.io/en/latest/

github: https://github.com/RandallBalestriero/SymJAX

pdf: https://arxiv.org/pdf/2005.10635v1.pdf

Evaluating Natural Language Generation with BLEURT

BLEURT (Bilingual Evaluation Understudy with Representations from Transformers)

https://ai.googleblog.com/2020/05/evaluating-natural-language-generation.html

Github: https://github.com/google-research/bleurt

Paper: https://arxiv.org/abs/2004.04696

Tensorflow vs PyTorch for Text Classification using GRU

https://medium.com/@rodolfosaldanha_71881/tensorflow-vs-pytorch-for-text-classification-using-gru-e95f1b68fa2d

CVML web lecture series on basics of deep learning. Registration required (June 3rd)
Deep Learning: Convolutional Neural Networks
Deep Object Detection

http://icarus.csd.auth.gr/cvml-web-lecture-series/

Andriy Burkov :
If you plan to do your Master's or Ph.D., choose your research advisor carefully. Ask about them his* current and past students, whether he was easily available for them, whether he responded to the requests for reference sent by their employers.

My research advisor, when was asked by one of my first employers whether they should hire me, responded: "You will call me back to thank me if you hire him." This is what a great research advisor would do for his students.

Unfortunately, not all research advisors are like this. For one of my hires, their research advisor didn't respond to two of my emails with requests for reference. I still hired the candidate, and I'm very happy with my choice. But the research advisor should be ashamed to ignore such requests about their recent alumni. To be available for their present and past students is advisors' direct duty.


🗣 @AI_Python_arXiv
✴️ @AI_Python_EN
❇️ @AI_Python

Adrian Olszewski :
EDIT: find the answer here: tinyurl.com/yc4m5lxx )

Dear Data Scientists. ANOVA is a powerful method. You often mention it in your posts. Sadly, I noticed, that you treat it mostly in the simplest way, while it's far beyond that! Well, Fisher didn't invent it with all those applications in mind, but it turned out, over time, that the procedure can be generalized greatly, constituting one of the most important methods in statistics.

You think you know all about it? ANOVA may surprise you.

You might have wondered, why:
- is ANOVA called in so many contexts: to compare means, models, testing contrasts?
- why is it called with either F or chi2 test (yes, it's about limiting distribution, but how?)
- why is it important to call it with appropriate type of sum of squares (when)?
- what is the relationship with LS-Means
- what does "joint test" actually means.

You might believe that the Tukey HSD method must agree with the result of F test in ANOVA. / No, it doesn't. Scheffe's does. /

If you pick the right answer, then do a research, you will understand how deep is the rabbit hole.

🗣 @AI_Python_arXiv
✴️ @AI_Python_EN
❇️ @AI_Python

NeuralPy

NeuralPy: A Keras like deep learning library works on top of PyTorch PyTorch is an open-source machine learning framework that accelerates the path from research prototyping to production deployment developed by Facebook runs on both CPU and GPU.

Github: https://github.com/imdeepmind/NeuralPy

Project: https://neuralpy.imdeepmind.com/

Acme: A new framework for distributed reinforcement learning by DeepMind
Intro:
https://deepmind.com/research/publications/Acme
Paper:
https://github.com/deepmind/acme/blob/master/paper.pdf
Repo:
https://github.com/deepmind/acme

#reinforcementlearning #ai #deepmind #deeplearning #machinelearning

🗣 @AI_Python_arXiv
✴️ @AI_Python_EN
❇️ @AI_Python

OpenAI Unveils 175 Billion Parameter GPT-3 Language Model
https://bit.ly/3dt4AkZ

🗣 @AI_Python_arXiv
✴️ @AI_Python_EN
❇️ @AI_Python

Interview with Professors in our Instagram

https://instagram.com/ai_python_podcast

MIT Deep Learning online course *New 2020 Edition*

For all lectures, slides, and lab materials

http://introtodeeplearning.com/

MIT Introduction to Deep Learning
Recurrent Neural Networks
Convolutional Neural Networks
Deep Generative Modeling
Reinforcement Learning
Deep Learning New Frontiers
Neurosymbolic AI
Generalizable Autonomy for Robot Manipulation
Neural Rendering
Machine Learning for Scent

Youtube videos:
https://www.youtube.com/watch?v=njKP3FqW3Sk&list=PLtBw6njQRU-rwp5__7C0oIVt26ZgjG9NI

🗣 @AI_Python_arXiv
✴️ @AI_Python_EN
❇️ @AI_Python

Reinforcement Learning

Let's say we have an agent in an unknown environment and this agent can obtain some rewards by interacting with the environment.

The agent is tasked to take actions so as to maximize cumulative rewards. In reality, the scenario could be a bot playing a game to achieve high scores, or a robot trying to complete physical tasks with physical items; and not just limited to these.

Like humans, RL agents learn for themselves to achieve successful strategies that lead to the greatest long-term rewards.

This kind of learning by trial-and-error, based on rewards or punishments, is known as reinforcement learning (RL).

TensorTrade is an open-source Python framework for building, training, evaluating, and deploying robust trading algorithms using reinforcement learning.

https://github.com/tensortrade-org/tensortrade

#artificialintelligence #machinelearning #datascience #datascience #python

🗣 @AI_Python_arXiv
✴️ @AI_Python_EN
❇️ @AI_Python

#TFHub tutorials TensorFlow Hub is a library for reusable machine learning modules : https://lnkd.in/eYk83J5 #DeepLearning #Tensorflow #TensorFlowHub #Tutorials

Big GANs Are Watching

You It is the state-of-the-art unsupervised GAN, which parameters are publicly available. They demonstrate that object saliency masks for GAN-produced images can be obtained automatically with BigBiGAN. These masks then are used to train a discriminative segmentation model. Being very simple and easy-to-reproduce, our approach provides competitive performance on common benchmarks in the unsupervised scenario.

Github: https://github.com/anvoynov/BigGANsAreWatching

Paper : https://arxiv.org/abs/2006.04988

#datascience #machinelearning #artificialintelligence #deeplearning

Google Research • Representation Learning for Information Extraction from Templatic Documents such as receipts, bills, insurance quotes. We propose a novel approach using representation learning for tackling the problem of extracting structured information from form-like document images.

Blogpost

https://ai.googleblog.com/2020/06/extracting-structured-data-from.html?m=1

Paper

https://research.google/pubs/pub49122/
We propose an extraction system that uses knowledge of the types of the target fields to generate extraction candidates, and a neural network architecture that learns a dense representation of each candidate based on neighboring words in the document. These learned representations are not only useful in solving the extraction task for unseen document templates from two different domains, but are also interpretable, as we show using loss cases. #machinelearning #deeplearning #datascience #dataengineer #nlp

The Most Important Fundamentals of PyTorch you Should Know

https://blog.exxactcorp.com/the-most-important-fundamentals-of-pytorch-you-should-know/

Code: https://github.com/tirthajyoti/PyTorch_Machine_Learning

Google Research • Representation Learning for Information Extraction from Templatic Documents such as receipts, bills, insurance quotes. We propose a novel approach using representation learning for tackling the problem of extracting structured information from form-like document images.

Blogpost

https://ai.googleblog.com/2020/06/extracting-structured-data-from.html?m=1

Paper

https://research.google/pubs/pub49122/
We propose an extraction system that uses knowledge of the types of the target fields to generate extraction candidates, and a neural network architecture that learns a dense representation of each candidate based on neighboring words in the document. These learned representations are not only useful in solving the extraction task for unseen document templates from two different domains, but are also interpretable, as we show using loss cases. #machinelearning #deeplearning #datascience #dataengineer #nlp