Object detection on aerial imagery using CenterNet. https://arxiv.org/abs/1908.08244

Progressive Face Super-Resolution via Attention to Facial Landmark. https://arxiv.org/abs/1908.08239

Boundary Aware Networks for Medical Image Segmentation. https://arxiv.org/abs/1908.08071

On-Device, Real-Time Hand Tracking with MediaPipe
Blog by Valentin Bazarevsky and Fan Zhang : https://ai.googleblog.com/2019/08/on-device-real-time-hand-tracking-with.html
#MachineLearning #TensorFlowLite #TensorFlow

Data Scientist (Machine Learning)

Experian DataLabs is a R&D unit at Experian formed with the desire to work in
collaboration with Experian’s business units to enhance relationships with clients and acquire strategic datasets. Experian® is a global leader in providing information, analytical tools and marketing services to organizations and consumers to help manage the risk and reward of commercial and financial decisions. Using our comprehensive understanding of individuals,
markets and economies, we help organizations find, develop and manage customer relationships to make their businesses more profitable.

Position Summary:
The person filling this position will be part of Experian North America R&D Data Lab concentrating on research and development of novel analytical solutions, new product prototyping, as well as new data asset evaluation and acquisition. This position requires extensive background and knowledge in machine learning and data science. A successful candidate should also have previous experience in developing deep-learning algorithms and/or deep learning based analytic solutions using large datasets. Experience in the areas of natural
language understanding, symbolic AI, online marketing, healthcare is a plus.
The candidate will need to be able to work on multiple concurrent projects, anticipate obstacles, and make high quality deliveries on an aggressive schedule. The candidate must also be a team player that is self-motivated and has excellent communication skills.
Key job functions include:
Analyzing, processing, evaluating and documenting large data sets
Identify/develop appropriate machine learning/deep learning/natural language understanding/natural language processing techniques to uncover the value of the data
Designing data structure and data storage schemes for efficient data manipulation and information retrieval
Developing tools for data processing and information retrieval
Developing data driven models to quantify the value of a given data set
Applying, modifying and inventing algorithms to solve challenging business problems
Validating score performance
Conducting ROI and benefit analysis
Documenting and presenting model process and model performance

https://ai-jobs.net/job/data-scientist-machine-learning-12/

A critique of pure learning and what artificial neural networks can learn from animal brains

In this Nature Communications Perspective, the author suggests that for AI to learn from animal brains, it is important to consider that animal behaviour results from brain connectivity specified in the genome through evolution, and not due to unique learning algorithms.

https://www.nature.com/articles/s41467-019-11786-6

PyTorch Implementation of Differentiable ODE Solvers
Differentiable ODE solvers with full GPU support and O(1)-memory backpropagation.
GitHub, by Ricky Chen : https://github.com/rtqichen/torchdiffeq
#Backpropagation #GPU #PyTorch

Write With Transformer
See how a modern neural network auto-completes your text 🤗
With the brand new GPT-2 large!
Built by the phenomenal Hugging Face team : https://transformer.huggingface.co
H / T : Lysandre Debut
#GPT2 #NeuralNetwork #Transformer

Top 10 Machine Learning Algorithms For Beginners
Linear Regression
Logistic regression
KNN Classification
Support Vector Machine (SVM)
Decision Trees
Random Forest
Artificial Neural Network
K-means Clustering
Naive Bayes theorem
Recurrent Neural Networks (RNN)
https://blog.quantinsti.com/top-10-machine-learning-algorithms-beginners/

Denmark: Researchers use machine learning to reveal gender stereotypes in literature https://www.technologynetworks.com/informatics/news/women-are-beautiful-men-are-rational-machine-learning-reveals-gender-stereotypes-in-literature-323254

Fusion of Detected Objects in Text for Visual Question Answering
Alberti et al.: https://arxiv.org/abs/1908.05054
#ArtificialIntelligence #DeepLearning #MachineLearning

The Generative Deep Learning Book
A great book, by David Foster : https://shop.oreilly.com/product/0636920189817.do
Blog : https://medium.com/applied-data-science/generative-deep-learning-the-parrot-has-landed-b291e6c254e
Thanks David for the signed copy !
#ArtificialIntelligence #DeepLearning #GenerativeDeepLearning

tf.keras for Researchers: Crash Course
"That's all you need to get started with reimplementing most deep learning research papers in TensorFlow 2.0 and Keras!"
Code by François Chollet: https://colab.research.google.com/drive/14CvUNTaX1OFHDfaKaaZzrBsvMfhCOHIR
Related very interesting new video : https://youtu.be/uhzGTijaw8A
#deeplearning #keras #tensorflow #tutorial

PaperRobot: Incremental Draft Generation of Scientific Ideas

Wang et al.: https://arxiv.org/abs/1905.07870

#ArtificialIntelligence #MachineLearning #DeepLearning

"Two neural nets learn to communicate through their own emergent visual language"
Here set in clay tablets.
By Joel Simon : https://github.com/joel-simon/dimensions-of-dialogue/blob/master/emergent_characters.ipynb
#language #neuralnetwork #deeplearning

On-Device, Real-Time Hand Tracking with MediaPipe
Blog by Valentin Bazarevsky and Fan Zhang : https://ai.googleblog.com/2019/08/on-device-real-time-hand-tracking-with.html
#MachineLearning #TensorFlowLite #TensorFlow