Science Courses and Tutorials

Google's recently released TensorFlow library has caused great waves in machine learning circles, with its powerful syntax that allows for distributed computation, improved efficiency, and modularisation. The framework allows you to build graph-based models, such as those used in machine learning and artificial intelligence, and have those models run on a distributed computing systems, including GPUs.

At this site, we have a collection of short tutorials designed to teach the concepts and usage to beginners. The goal is to introduce TensorFlow without the data mining component, allowing you to focus on learning the framework and not needing to learn about data mining techniques at the same time.

This 3-hour course (video + slides) offers developers a quick introduction to deep-learning fundamentals, with some TensorFlow thrown into the bargain.

Deep learning (aka neural networks) is a popular approach to building machine-learning models that is capturing developer imagination. If you want to acquire deep-learning skills but lack the time, I feel your pain.

I thought that the results from pix2pix by Isola et al. looked pretty cool and wanted to implement an adversarial net, so I ported the Torch code to Tensorflow. The single-file implementation is available as pix2pix-tensorflow on github.

Here are some examples of what this thing does, from the original paper.

This Python notebook implements the toy example from Adversarial Variational Bayes: Unifying Variational Autoencoders and Generative Adversarial Networks by Lars Mescheder, Sebastian Nowozin, Andreas Geiger.

An article and tutorial on using linters to manage large, enterprise-sized Python projects.

A long compilation of d3 tutorials for data visualization are hosted on github, from core concepts to specialized techniques and applications. Also, links to several books, blogs, slides, courses, and lecture videos are listed.

Slides from a talk given by Jeff Dean from the Google Brain team.

The method is called after the city in the Monaco principality, because of a roulette, a simple random number generator. The name and the systematic development of Monte Carlo methods dates from about 1944....

Try a one-hour tutorial designed for all ages in over 45 languages. Join millions of students and teachers in over 180 countries starting with an Hour of Code.

In this course we aim to teach you how to think critically about the data and models that constitute evidence in the social and natural sciences.

Our learning objectives are straightforward. After taking the course, you should be able to:

- Remain vigilant for bullshit contaminating your information diet.
- Recognize said bullshit whenever and wherever you encounter it.
- Figure out for yourself precisely why a particular bit of bullshit is bullshit.
- Provide a statistician or fellow scientist with a technical explanation of why a claim is bullshit.
- Provide your crystals-and-homeopathy aunt or casually racist uncle with an accessible and persuasive explanation of why a claim is bullshit.

We will be astonished if these skills do not turn out to be among the most useful and most broadly applicable of those that you acquire during the course of your college education.

Taught by Carl T. Bergstrom, UW Department of Biology, and Jevin West, UW Information School.

This site uses the tools of modern economics and game theory to explore how the interaction of intelligent goal-seeking individuals determines social outcomes. By David K. Levine, Department of Economics and Robert Schuman Center for Advanced Study Joint Chair at the European University Institute. He is also John H. Biggs Distinguished Professor of Economics Emeritus at Washington University in St. Louis.

Jupyter (formerly known as IPython) notebooks are great – but have you ever accidentally deleted a cell that contained a really important function that you want to keep? Well, this post might help you get it back.

Changes in technology---specifically the transition from the analog age to the digital age---mean that we can now collect and analyze social data in new ways. This six week mini-class is about doing social research in these new ways. Unlike some other courses on computational social science, this course will emphasize "social science" and de-emphasize "computation." We will focus on how traditional concepts of research design in the social sciences can inform our understanding of new data sources, and how these new data sources might require us to update our thinking on research design. The course should be helpful for social scientists that want to do more data science and data scientists that want to do more social science. Public course materials are on github: https://github.com/msalganik/soc596_f2016.

The VEX Robotics system allows students and hobbyists to build "real" robots. VEX kits provide the whole solution or can be integrated with large and small motors, pneumatic actuators, sensors, electronic components, composites, and a student's imagination to engineer a unique mechatronic solution. Designed by scientists and educators at the Carnegie Mellon Robotics Academy.

A few months ago, one of my friends asked me if I could help him extract some data from a collection of PDFs. The PDFs contained records of his financial transactions over a period of years and he wanted to analyze them. Unfortunately, Excel and plain text versions of the files were no longer available, so the PDFs were his only option.

I reviewed a few Python-based PDF parsers and decided to try Tika, which is a port of Apache Tika. Tika parsed the PDFs quickly and accurately. I extracted the data my friend needed and sent it to him in CSV format so he could analyze it with the program of his choice. Tika was so fast and easy to use that I really enjoyed the experience. I enjoyed it so much I decided to write a blog post about parsing PDFs with Tika.

The University of California at Berkeley has posted course webcasts and podcasts on a single page for easy access to courses in all fields from art to science to liberal arts. Links to course syllibi and course schedules are also available.

Dask Dataframe extends the popular Pandas Python library to operate on big data-sets on a distributed cluster. We show its capabilities by running through common dataframe operations on a common dataset.

Air pollution can make you sick. Air pollution can damage the environment. Air pollution can damage property. Air pollution can cause haze, reducing visibility in national parks and sometime interfering with aviation. The Clean Air Act will improve air quality in the United States, a good thing for your health, your property and the environment. The 1990 Clean Air Act is lengthy--about 800 pages--because it tackles many difficult and complicated air pollution problems. We have prepared this summary of the 1990 version of the Clean Air Act because we think everyone should understand what is in the law and how it may affect them.

An explanation of the Riemann Hypothesis by Jorgen Veisdal.

This is a ArXiv report summarizing the tutorial presented by Ian Goodfellow at NIPS 2016 on generative adversarial networks (GANs). The tutorial describes: (1) Why generative modeling is a topic worth studying, (2) how generative models work, and how GANs compare to other generative models, (3) the details of how GANs work, (4) research frontiers in GANs, and (5) state-of-the-art image models that combine GANs with other methods. Finally, the tutorial contains three exercises for readers to complete, and the solutions to these exercises.