Federated Learning: Collaborative Machine Learning without Centralized Training Data

https://ai.googleblog.com/2017/04/federated-learning-collaborative.html

Glow is a machine learning compiler and execution engine for hardware accelerators. It is designed to be used as a backend for high-level machine learning frameworks. The compiler is designed to allow state of the art compiler optimizations and code generation of neural network graphs

https://github.com/pytorch/glow

Papers With Code now includes 950+ ML tasks, 500+ evaluation tables (including SOTA results) and 8500+ papers with code.

https://paperswithcode.com

لینک جدید گروه
https://xn--r1a.website/joinchat/A3HTSj3_zWPMCwcByv1aKg

Training deep learning models with vast amounts of data is necessary to achieve accurate results. Data in the wild, or even prepared data sets, is usually not in the form that can be directly fed into neural network. This is where NVIDIA DALI data preprocessing comes into play. DALI is a set of highly optimized building blocks plus an execution engine to accelerate input data pre-processing for deep learning applications.
https://devblogs.nvidia.com/fast-ai-data-preprocessing-with-nvidia-dali/

بینایی انسان و ماشین
سمپوزیوم علوم اعصاب دانشگاه صنعتی شریف

- آشنایی با سیستم بینایی انسان
- پردازش تصاویر دیجیتال و بینایی ماشین
- آشنایی با تصویر برداری پزشکی و کاربردها
- آشنایی با پردازش تصاویر پزشکی

🗓 تاریخ برگزاری : پنجشنبه 2 اسفند
💶 هزینه ثبت نام : 150 هزار تومان
🎓 اعطای مدرک معتبر دانشگاه صنعتی شریف
⚠️ ظرفیت ثبت نام محدود

https://pedu.sharif.edu/events/details/1034

#اسفند1397
#Workshop #Machine_Vision #Machine_Learning #Python #OpenCV #TensorFlow #SNS #FK
#Tehran #SUT
sns.ee.sharif.edu
https://xn--r1a.website/convent/3445

در حال حاضر خلاصه کلاس CS ۲۳۰ - یادگیری عمیق استنفورد به زبان فارس در دسترس است

راهنمای کوتاه نکات و ترفندهای یادگیری عمیق

https://stanford.edu/~shervine/l/fa/teaching/cs-230/cheatsheet-deep-learning-tips-and-tricks

راهنمای کوتاه شبکه‌های عصبی پیچشی (کانولوشنی)

https://stanford.edu/~shervine/l/fa/teaching/cs-230/cheatsheet-convolutional-neural-networks

راهنمای کوتاه شبکه‌های عصبی برگشتی

https://stanford.edu/~shervine/l/fa/teaching/cs-230/cheatsheet-recurrent-neural-networks

Notebooks for Neural Processes (NPs) and Attentive Neural Processes (ANPs)
https://github.com/deepmind/neural-processes/blob/master/attentive_neural_process.ipynb

PlotNeuralNet
Latex code for drawing neural networks for reports and presentation. Have a look into examples to see how they are made. Additionally, lets consolidate any improvements that you make and fix any bugs to help more people with this code.
https://github.com/HarisIqbal88/PlotNeuralNet

Seven Myths in Machine Learning Research
Myth1: TensorFlow is a Tensor manipulation library
Myth 2: Image datasets are representative of real images found in the wild
Myth 3: Machine Learning researchers do not use the test set for validation
Myth 4: Every datapoint is used in training a neural network
Myth 5: We need (batch) normalization to train very deep residual networks
Myth 6: Attention > Convolution
Myth 7: Saliency maps are robust ways to interpret neural networks


https://crazyoscarchang.github.io/2019/02/16/seven-myths-in-machine-learning-research/

Google AI Blog: Introducing GPipe, an Open Source Library for Efficiently Training Large-scale Neural Network Models
http://ai.googleblog.com/2019/03/introducing-gpipe-open-source-library.html

Feature visualizations, letting us see through the eyes of the neural network:
https://distill.pub/2019/activation-atlas/

#TensorFlow Lite running on a Cortex M4 developer board, handling simple speech keyword recognition
https://petewarden.com/2019/03/07/launching-tensorflow-lite-for-microcontrollers/

A step-by-step #tutorial showing how to perform #FederatedLearning using the same infrastructure #Google uses on 10s of millions of smartphones with #TensorFlow

https://medium.com/tensorflow/introducing-tensorflow-federated-a4147aa20041

#آموزش #ویدیو #سورس_کد

اسلایدها:
https://www.slideshare.net/Alirezaakhavanpour/deep-face-recognition-oneshot-learning
تشخیص چهره- بخش اول- علیرضا اخوان پور-1

•one-shot learning: Face Verification & Recognition
•Siamese network
•facenet triplet loss
https://www.aparat.com/v/xdC7r
تشخیص چهره - علیرضا اخوان پور - بخش دوم
• one-shot learning: Face Verification & Recognition
• Discriminative Feature
• Center loss
https://www.aparat.com/v/qmo3u

کدهای این ارائه
در گیت هاب هوش پارت و همچنین آدرس گیت هاب زیر موجود است:

https://github.com/Alireza-Akhavan/deep-face-recognition

🙏Thanks to: @partdpai
#face #face_recognition #verification

Discover Computer Vision Datasets:

https://www.visualdata.io

A new summary of The Evolved #Transformer
The recipe:
1. Take the human-designed Transformer
2. Add a new neural architecture search - PDH
3. Whisk with ~200 TPUs to get a much better model - #ET

https://www.lyrn.ai/2019/03/12/the-evolved-transformer/

Rich Sutton the Bitter Lesson

http://www.incompleteideas.net/IncIdeas/BitterLesson.html

Whiteson, who strongly disagrees with Sutton's point of view, believes that the history of AI teaches us that leveraging computation always eventually wins out over leveraging human knowledge.


Sutton says that the intrinsic complexity of the world means we shouldn’t build prior knowledge into our systems. But I conclude the exact opposite: that complexity leads to crippling intractability for the search and learning approaches on which Sutton proposes to rely.

https://twitter.com/shimon8282/status/1106534178676506624?s=19

پنج لکچر ابتدایی کورس cs224n استنفورد -زمستان ۲۰۱۹ (پردازش زبان طبیعی) منتشر شد.
ویدئوهای بعدی نیز تا زمان اتمام رسمی این کورس منتشر خواهند شد

https://www.youtube.com/playlist?list=PLoROMvodv4rOhcuXMZkNm7j3fVwBBY42z