Query2vec: Search query expansion with query embeddings
https://bytes.grubhub.com/search-query-embeddings-using-query2vec-f5931df27d79

The Measure of Intelligence
François Chollet : https://arxiv.org/abs/1911.01547
GitHub : https://github.com/fchollet/ARC
#ArtificialIntelligence #DeepLearning #MachineLearning

Dynamics-Aware Unsupervised Discovery of Skills
Sharma et al.: https://arxiv.org/abs/1907.01657
#MachineLearning #Robotics #ReinforcementLearning

Deep learning reveals cancer metastasis and therapeutic antibody targeting in whole body
https://www.biorxiv.org/content/biorxiv/early/2019/02/05/541862.full.pdf

A GPT-2 style model for dialog
A State-of-the-Art Large-scale Pretrained Response generation model (DialoGPT)
code https://github.com/microsoft/DialoGPT
paper https://arxiv.org/pdf/1911.00536.pdf

AI meets physics - using artificial neural networks to approximate solutions of the three-body problem.


I'm increasingly intrigued by this paper (https://arxiv.org/pdf/1910.07291.pdf) showing the application of Artificial Neural networks to the infamously insoluble three-body problem in physics, where we try to work out the future position of three objects sometime in the future given Newton's equations of motion. I think it has important implications to how we think about approximation and how we achieve it in practice.

From the authors: "Our results provide evidence that, for computationally challenging regions of phase-space, a trained ANN can replace existing numerical solvers, enabling fast and scalable simulations of many-body systems to shed light on outstanding phenomena such as the formation of black-hole binary systems or the origin of the core collapse in dense star clusters."

https://t.iss.one/ArtificialIntelligenceArticles

5 challenges for the next 5 years of computer vision research by Jitendra Malik at ICCV2019

The latest from TensorFlow
Tensorflow 2.0
Transformers library
Up to 3x training performance improvement
Addons and extensions
Tensorboard, debugging and visualization
Tensorflow Hub: pretrained models
Deploy ML anywhere: TF-extended (server), TF-lite (mobile) and TF-js (web)

https://www.youtube.com/watch?v=n56syJSLouA

CE7454 : Deep Learning for Data Science
Lecture 13: Attention Neural Networks
Xavier Bresson : https://dropbox.com/s/kbrsvhwe2lac1uo/lecture13_attention_neural_networks.pdf?dl=0
Demo :
https://github.com/xbresson/CE7454_2019/blob/master/codes/labs_lecture13/seq2seq_transformers_demo.ipynb
#DeepLearning #DataScience #Transformer

Building the first holographic brain 'atlas'

A team of researchers, led by Case Western Reserve University scientists and technicians using the Microsoft HoloLens mixed reality platform, has created what is believed to be the first interactive holographic mapping system of the axonal pathways in the human brain.

https://medicalxpress.com/news/2019-11-holographic-brain-atlas.html

Fruit identification using Arduino and TensorFlow

By Dominic Pajak and Sandeep Mistry : https://blog.arduino.cc/2019/11/07/fruit-identification-using-arduino-and-tensorflow/

#Arduino #TensorFlow #DeepLearning

Graph Neural Tangent Kernel: Fusing Graph Neural Networks with Graph Kernels #NeurIPS2019
Du et al. : https://arxiv.org/abs/1905.13192
GitHub : https://github.com/KangchengHou/gntk
#MachineLearning #ArtificialIntelligence #DeepLearning

Feedforward and feedback processes in visual recognition
https://www.youtube.com/watch?v=a4yoqdUr2hU

HoloGAN (A new generative model) learns 3D representation from natural images

Paper: https://arxiv.org/pdf/1904.01326.pdf
Github: https://github.com/thunguyenphuoc/HoloGAN
Dataset: https://mmlab.ie.cuhk.edu.hk/projects/CelebA.html
https://www.marktechpost.com/2019/11/04/hologan-a-new-generative-model-learns-3d-representation-from-natural-images/

Deep Learning for Computational Chemistry

Garrett B. Goh, Nathan Oken Hodas, Abhinav Vishnu

Published in Journal of Computational… 2017

DOI:10.1002/jcc.24764

Arxiv Free Download:
https://arxiv.org/abs/1701.04503

Paywall:
https://onlinelibrary.wiley.com/doi/abs/10.1002/jcc.24764

#deeplearning #AI #artificialintelligence #chemistry #computationalchemistry

In this review, we provide an introductory overview into the theory of deep neural networks and their unique properties that distinguish them from traditional machine learning algorithms used in cheminformatics.

By providing an overview of the variety of emerging applications of deep neural networks, we highlight its ubiquity and broad applicability to a wide range of challenges in the field, including quantitative structure activity relationship, virtual screening, protein structure prediction, quantum chemistry, materials design, and property prediction.

In reviewing the performance of deep neural networks, we observed a consistent outperformance against non-neural networks state-of-the-art models across disparate research topics, and deep neural network-based models often exceeded the "glass ceiling" expectations of their respective tasks.