This was posted as a question on StackExchange. The state of the art of non-linearity is to use rectified linear units (ReLU) instead of sigmoid function in deep neural network. What are the advantages? I know that training a network when ReLU is used would be faster, and it is more biological inspired, what are the other advantages? (That is, any disadvantages of using sigmoid)?…

Recently Google DeepMind announced AlphaGo Zero — an extraordinary achievement that has shown how it is possible to train an agent to a superhuman level in the highly complex and challenging domain of Go, ‘tabula rasa’ — that is, from a blank slate, with no human expert play used as training data.

It thrashed the previous reincarnation 100–0, using only 4TPUs instead of 48TPUs and a single…

Guest blog by Jim Frost.

Regression analysis mathematically…

Today’s Web-enabled deluge of electronic data calls for automated methods of data analysis. Machine learning provides these, developing methods that can automatically detect patterns in data and then use the uncovered patterns to predict future data. This textbook offers a comprehensive and self-contained introduction to the field of machine learning, based on a unified, probabilistic approach. 



The coverage combines breadth and depth, offering necessary…

Handbook of Statistical Analysis and Data Mining Applications, Second Edition, is a comprehensive professional reference book that guides business analysts, scientists, engineers and researchers, both academic and industrial, through all stages of data analysis, model building and implementation. The handbook helps users discern technical and business problems, understand the strengths and weaknesses of modern data mining algorithms and employ the right statistical methods for…

Yet another one of these One Picture tutorials, and in some ways, in the same old-fashioned style as our Type I versus Type II Errors in One Picture.…

This is Google's dictionary on machine learning. Very interesting, with more than 100 entries, but clearly biased towards what Google thinks machine learning is. There is still big room for improvement, as this glossary is missing many important entries such as

• maximum likelihood
• Bayesian networks
• dimension reduction
• hierarchical modeling
• survival analysis
• Markov property
• cross-validation
• time…

Which type are you? Can you recognize the programming language used in this illustration? Click on the picture to zoom in. …

The format is very similar to a BIG cheat sheet. This cookbook integrates a variety of topics in probability theory and statistics. It is based on literature and in-class material from courses of the statistics department at the University of California in Berkeley but also influenced by other sources . 

Author: Matthias Vallentin…

Guest blog by Rob Kabacoff. Rob is Professor of Quantitative Analytics at Wesleyan University.

R is an elegant and comprehensive statistical and graphical programming language. Unfortunately, it can also have a steep learning curve. I created this website for both current R users, and experienced users of other statistical packages (e.g., SAS, SPSS, Stata) who…

Interesting infographics produced by PwC. To view the original article, download the infographics in PDF format, and read the comments, click here. 

DSC Resources

• Services: Hire a Data Scientist | …

From the author of the bestsellers, Data Scientist and Julia for Data Science, this book covers four foundational areas of data science. The first area is the data science pipeline including methodologies and the data scientist's toolbox. The second are essential practices needed in understanding the data including questions and hypotheses. The third are pitfalls to avoid in the data science process. The fourth is an awareness of future trends…

Learn how to use R to turn raw data into insight, knowledge, and understanding. This book introduces you to R, RStudio, and the tidyverse, a collection of R packages designed to work together to make data science fast, fluent, and fun. Suitable for readers with no previous programming experience, R for Data Science is designed to get you doing data science as quickly as possible.…

Below is an extract from a 36-page report entitled "Technology and Innovation for the Future of Production: Accelerating Value Creation", available for free here, and produced by the World Economic Forum.

The extract below, about the future of AI, is figure 7 at page 13. This long report also discusses other interested topics and is peppered with many useful…

The question was posted on Quora as "What do you look for when hiring an entry-level data scientist? Would a master’s in Data Science or a bootcamp be beneficial?" The answer below is from Eduardo Arino de la Rubia, Chief Data Scientist at Domino Data Lab.

I think that realistically, either one of them is enough to get me to…

This cheat sheet was produced by DataCamp, and it is based on the Keras library..Keras is an easy-to-use and powerful library for Theano and TensorFlow that provides a high-level neural networks API to develop and evaluate deep learning models. Originally posted here in PDF format. Click on the image below to zoom in. …

This infographic was produced by Springboard, and it lists a few short online, inexpensive courses along with some university programs, leading to a certificate. The infographics provides some highlights for each program, for comparison purposes. For more data science programs and certificates, click here or…

By Winnifred Louis, Associate Professor, Social Psychology, The University of Queensland, and Cassandra Chapman,PhD Candidate in Social Psychology, The University of Queensland.…

This Python cheat sheet was produced by DataCamp. Click on the image to zoom in. The original, published here, is available as a PDF document. The translation from PDF to image format was done using PDF2PNG. 

To check dozens of data science related cheat sheets,…

This is a curated list of the most cited deep learning papers (since 2012) posted by Terry Taewoong Um.

Source for picture: What is deep…