Official information of Coronavirus disease 2019 (COVID-19) in South Korea is available in Kaggle
https://www.kaggle.com/kimjihoo/coronavirusdataset
https://www.kaggle.com/kimjihoo/coronavirusdataset
Kaggle
[NeurIPS 2020] Data Science for COVID-19 (DS4C)
DS4C: Data Science for COVID-19 in South Korea
Postdoc position in ML/stats at the University of Chicago
Applications are invited for a postdoctoral researcher working under the supervision of Prof. Bryon Aragam in statistics, machine learning, and/or optimization within the Econometrics and Statistics group at the Booth School of Business of the University of Chicago. Potential candidates should have a background in statistics and machine learning, for example nonconvex optimization, nonparametric statistics, and/or learning theory. Applications of interest include causal inference, representation learning, personalization, and graphical models, but may also depend on the candidate’s individual research interests. There will be an emphasis on theoretical/mathematical problems as well as computational/applied work, with a particular focus on problems at the intersection.
The Econometrics and Statistics group at the University of Chicago is diverse and rapidly growing, with 12 full-time faculty working in diverse areas such as statistical machine learning, causal inference, Bayesian statistics, financial econometrics, and forecasting.
Qualifications
The candidate should have a recent Ph.D. degree (or all-but-dissertation) in statistics, computer science, mathematics, or a related area, and should be proficient in programming in Python or R. We welcome applications from candidates with diverse and/or nontraditional backgrounds.
To Apply
Interested candidates should email bryon@chicagobooth.edu to indicate their interest in this position.
Required Documents
1) Resume/CV
2) Cover letter, including brief denoscription of research interests
3) Graduate trannoscripts
4) At least one academic reference
Application link: https://uchicago.wd5.myworkdayjobs.com/en-US/External/job/Hyde-Park-Campus/Principal-Researcher_JR07406
Applications are invited for a postdoctoral researcher working under the supervision of Prof. Bryon Aragam in statistics, machine learning, and/or optimization within the Econometrics and Statistics group at the Booth School of Business of the University of Chicago. Potential candidates should have a background in statistics and machine learning, for example nonconvex optimization, nonparametric statistics, and/or learning theory. Applications of interest include causal inference, representation learning, personalization, and graphical models, but may also depend on the candidate’s individual research interests. There will be an emphasis on theoretical/mathematical problems as well as computational/applied work, with a particular focus on problems at the intersection.
The Econometrics and Statistics group at the University of Chicago is diverse and rapidly growing, with 12 full-time faculty working in diverse areas such as statistical machine learning, causal inference, Bayesian statistics, financial econometrics, and forecasting.
Qualifications
The candidate should have a recent Ph.D. degree (or all-but-dissertation) in statistics, computer science, mathematics, or a related area, and should be proficient in programming in Python or R. We welcome applications from candidates with diverse and/or nontraditional backgrounds.
To Apply
Interested candidates should email bryon@chicagobooth.edu to indicate their interest in this position.
Required Documents
1) Resume/CV
2) Cover letter, including brief denoscription of research interests
3) Graduate trannoscripts
4) At least one academic reference
Application link: https://uchicago.wd5.myworkdayjobs.com/en-US/External/job/Hyde-Park-Campus/Principal-Researcher_JR07406
Reduce model size to train/test faster.
However, you should actually increase model size to speed up training and inference for transformers
Speeding Up Transformer Training and Inference By Increasing Model Size
https://bair.berkeley.edu/blog/2020/03/05/compress/
paper https://arxiv.org/pdf/2002.11794.pdf
However, you should actually increase model size to speed up training and inference for transformers
Speeding Up Transformer Training and Inference By Increasing Model Size
https://bair.berkeley.edu/blog/2020/03/05/compress/
paper https://arxiv.org/pdf/2002.11794.pdf
The Berkeley Artificial Intelligence Research Blog
Speeding Up Transformer Training and Inference By <i>Increasing</i> Model Size
The BAIR Blog
How a research scientist built Kornia: an open source differentiable library for PyTorch
https://medium.com/pytorch/how-a-research-scientist-built-kornia-an-open-source-differentiable-library-for-pytorch-16e2aa758bc8
https://medium.com/pytorch/how-a-research-scientist-built-kornia-an-open-source-differentiable-library-for-pytorch-16e2aa758bc8
Medium
How a research scientist built Kornia: an open source differentiable library for PyTorch
This is a blog post covering an interview with Edgar Riba, the founding member of Kornia, an open source differentiable CV library.
Diet modulates brain network stability, a biomarker for brain aging, in young adults
https://www.pnas.org/content/early/2020/03/02/1913042117.long
https://www.pnas.org/content/early/2020/03/02/1913042117.long
PNAS
Diet modulates brain network stability, a biomarker for brain aging, in young adults
To better understand how diet influences brain aging, we focus here on the presymptomatic period during which prevention may be most effective. Large-scale life span neuroimaging datasets show functional communication between brain regions destabilizes with…
INTERVIEW WITH JUERGEN SCHMIDHUBER
https://analyticsindiamag.com/lstm-juergen-schmidhuber-artificial-intelligence-switzerland-goedel-machines/
https://analyticsindiamag.com/lstm-juergen-schmidhuber-artificial-intelligence-switzerland-goedel-machines/
Analytics India Magazine
Interview With Juergen Schmidhuber
Juergen Schmidhuber believes that this decade will witness proliferation of Active AI in industrial processes and machines and robots.
Nonlinear neural network dynamics accounts for human confidence in a sequence of perceptual decisions
https://www.biorxiv.org/content/10.1101/648022v3
https://www.biorxiv.org/content/10.1101/648022v3
bioRxiv
NONLINEAR NEURAL NETWORK DYNAMICS ACCOUNTS FOR HUMAN CONFIDENCE IN A SEQUENCE OF PERCEPTUAL DECISIONS
Electrophysiological recordings during perceptual decision tasks in monkeys suggest that the degree of confidence in a decision is based on a simple neural signal produced by the neural decision process. Attractor neural networks provide an appropriate biophysical…
Berkeley's CS294-158 videos is an excellent class on Deep Unsupervised Learning
https://sites.google.com/view/berkeley-cs294-158-sp20/home
https://sites.google.com/view/berkeley-cs294-158-sp20/home
Google
CS294-158-SP20 Deep Unsupervised Learning Spring 2020
About: This course will cover two areas of deep learning in which labeled data is not required: Deep Generative Models and Self-supervised Learning. Recent advances in generative models have made it possible to realistically model high-dimensional raw data…
PRML algorithms implemented in Python
Python codes implementing algorithms described in Bishop's book "Pattern Recognition and Machine Learning" : https://github.com/ctgk/PRML
#DeepLearning #MachineLearning #Python
Python codes implementing algorithms described in Bishop's book "Pattern Recognition and Machine Learning" : https://github.com/ctgk/PRML
#DeepLearning #MachineLearning #Python
GitHub
GitHub - ctgk/PRML: PRML algorithms implemented in Python
PRML algorithms implemented in Python. Contribute to ctgk/PRML development by creating an account on GitHub.
Understanding Self-Training for Gradual Domain Adaptation
Kumar et al.: https://arxiv.org/abs/2002.11361
#ArtificialIntelligence #DeepLearning #MachineLearning
Kumar et al.: https://arxiv.org/abs/2002.11361
#ArtificialIntelligence #DeepLearning #MachineLearning
Artificial intelligence has been used for the first time to instantly and accurately measure blood flow, in a study led by UCL and Barts Health NHS Trust
Image: Myocardial blood flow "perfusion map" created and analysed using AI showing an area of the heart receiving a reduced blood supply (arrow) and putting the patient at risk of heart attacks and other adverse events.
https://www.ucl.ac.uk/news/2020/feb/ai-helps-predict-heart-attacks-and-stroke
Image: Myocardial blood flow "perfusion map" created and analysed using AI showing an area of the heart receiving a reduced blood supply (arrow) and putting the patient at risk of heart attacks and other adverse events.
https://www.ucl.ac.uk/news/2020/feb/ai-helps-predict-heart-attacks-and-stroke
UCL News
AI helps predict heart attacks and stroke
Artificial intelligence has been used for the first time to instantly and accurately measure blood flow, in a study led by UCL and Barts Health NHS Trust.