JOB: postdoc in deep learning with satellite images for global health


We are seeking outstanding candidates with expertise in image processing, deep learning and machine learning. The ideal candidate is an expert in convolutional neural networks and generative models such as GANs and has experience programming statistical tools in TensorFlow. The post will involve image processing on satellite imagery to predict health outcomes.
Full job posting: https://www.imperial.ac.uk/jobs/description/MED01288/research-associate/

Closing date: 11 July 2019

For informal enquiries, contact Seth Flaxman <[email protected]> or Samir Bhatt <[email protected]>

--
Dr Seth Flaxman
Lecturer, Machine Learning and Big Data Analytics
Department of Mathematics and Data Science Institute
Imperial College London
www.sethrf.com

Loving this step-by-step guide to Kernels by Abdul Majed Raja https://towardsdatascience.com/kaggle-kernels-for-beginners-a-step-by-step-guide-3db6b1cd7606

MeetUp! A Corpus of Joint Activity Dialogues in a Visual Environment
https://arxiv.org/abs/1907.05084

Understanding the Representation Power of Graph Neural Networks in Learning Graph Topology. arxiv.org/abs/1907.05008

Representing text in natural language processing
https://towardsdatascience.com/representing-text-in-natural-language-processing-1eead30e57d8

Why do Neural Networks Need an Activation Function?
https://www.datastuff.tech/machine-learning/why-do-neural-networks-need-an-activation-function/

BigGAN + BiGAN = BigBiGAN: New State-of-the art Model In Representation Learning
https://neurohive.io/en/news/biggan-bigan-bigbigan-new-state-of-the-art-model-in-representation-learning/

Making AI Forget You: Data Deletion in Machine Learning. arxiv.org/abs/1907.05012

Talos
Hyperparameter Optimization for Keras: https://github.com/autonomio/talos
#DeepLearning #Keras #TensorFlow #MachineLearning

Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm
Silver et al.: https://arxiv.org/abs/1712.01815
#artificialintelligence #chess #reinforcementlearning

This paper reveals the potential vulnerabilities of LiDAR-based autonomous driving detection systems, by proposing an optimization based approach LiDAR-Adv.

https://www.profillic.com/paper/arxiv:1907.05418

This approach is used to generate adversarial objects that can evade the LiDAR-based detection system under various conditions.

Dopamine: a research framework for fast prototyping of reinforcement learning algorithms

Code: https://github.com/google/dopamine

#ArtificialIntelligence #DeepLearning #ReinforcementLearning

How to Code the GAN Training Algorithm and Loss Functions
https://machinelearningmastery.com/how-to-code-the-generative-adversarial-network-training-algorithm-and-loss-functions/

Deep learning based super resolution, without using a GAN
Blog by Christopher Thomas: https://towardsdatascience.com/deep-learning-based-super-resolution-without-using-a-gan-11c9bb5b6cd5
#MachineLearning #DeepLearning #ArtificialIntelligence

WikiMatrix: Mining 135M Parallel Sentences in 1620 Language Pairs from Wikipedia
Schwenk et al.: https://arxiv.org/abs/1907.05791
Data: https://github.com/facebookresearch/LASER/tree/master/tasks/WikiMatrix
#artificialintelligence #bigdata #datascience #machinelearning

AI Habitat tensorboard integration
Watch videos of your agents in training!
Courtesy: Jason Jiazhi Zhang: https://github.com/facebookresearch/habitat-api/pull/127
Habitat-API, by Facebook Research, is a modular high-level library to train embodied AI agents across a variety of tasks, environments, and simulators: https://aihabitat.org
#ArtificielIntelligence #DeepLearning #ReinforcementLearning

NPA: Neural News Recommendation with Personalized Attention. arxiv.org/abs/1907.05559