Science Blogs

My name is Mercedes Gimeno-Segovia, I am a quantum physicist whose research mainly focuses on large-scale photonic quantum computing, developing new theoretical ideas and protocols that can be easily implemented in the lab.

t-SNE is the very popular algorithm to extremely reduce the dimensionality of your data in order to visually present it. It is capable of mapping hundreds of dimensions to just 2 while preserving important data relationships, that is, when closer samples in the original space are closer in the reduced space. t-SNE works quite well for small and moderately sized real-world datasets and does not require much tuning of its hyperparameters. In other words, if you’ve got less than 100,000 points, you will apply that magic black box thing and get a beautiful scatter plot in return.

In New York state, neighbors are testing their ability to sell solar energy to one another using blockchain technology. In Austria, the country’s largest utility conglomerate, Wien Energie, is taking part in a blockchain trial focused on energy trading with two other utilities. Meanwhile in Germany, the power company Innogy is running a pilot to see if blockchain technology can authenticate and manage the billing process for autonomous electric-vehicle charging stations.

Blockchain has grabbed the attention of the heavily regulated power industry as it braces for an energy revolution in which both utilities and consumers will produce and sell electricity.

This is an open science weblog focused on quantum information theory, condensed matter physics, and mathematical physics.

I am a researcher in quantum information theory based at the Institut für Theoretische Physik, Leibniz Universität Hannover. I am interested in quantum information theory and complex quantum systems.

Inspired by the recent discussions (see, eg., Michael Nielsen’s blog postings) surrounding open science I decided to start a weblog containing my research notes. I intend to post my notes on my various research projects along with some expository material. It is my hope that this weblog can become a forum for open discussion and research. Thus I would like to invite you to comment, contribute, and collaborate! If you make any progress on the ideas or open problems discussed on this site then please let me know via the comments: I am more than happy for anything here to become a joint project, the more the merrier!

Deep learning is a field with intense computational requirements and the choice of your GPU will fundamentally determine your deep learning experience. With no GPU this might look like months of waiting for an experiment to finish, or running an experiment for a day or more only to see that the chosen parameters were off. With a good, solid GPU, one can quickly iterate over deep learning networks, and run experiments in days instead of months, hours instead of days, minutes instead of hours. So making the right choice when it comes to buying a GPU is critical. So how do you select the GPU which is right for you? This blog post will delve into that question and will lend you advice which will help you to make choice that is right for you.

Here is a simple HowTo to understand the concept of shapes in TensorFlow and hopefully avoid losing hours of debugging them.

I recently finished Walter Isaacson’s biography of Albert Einstein. The biography covers not just Einstein’s intellectual achievements, but also his anti-war activism, marital difficulties and celebrity. However, I wanted to share just the one part I found most interesting: how did Einstein learn?

The Earth's climate has changed throughout history. Just in the last 650,000 years there have been seven cycles of glacial advance and retreat, with the abrupt end of the last ice age about 7,000 years ago marking the beginning of the modern climate era — and of human civilization. Most of these climate changes are attributed to very small variations in Earth’s orbit that change the amount of solar energy our planet receives.

This is the blog of California Water Research, which collaborates with a wide range of environmental, fishing, and local groups on research and analysis of California’s State Water Project and Central Valley Project, and agricultural and urban water use in California.

I’ve recently given a couple of talks (PyGotham video, PyGotham slides, Strata NYC slides) about text summarization.

I cover three ways of automatically summarizing text. One is an extremely simple algorithm from the 1950s, one uses Latent Dirichlet Allocation, and one uses skipthoughts and recurrent neural networks.

The talk is conceptual, and avoids code and mathematics. So here is a list of resources if you’re interested in text summarization and want to dive deeper.

Given all the hype and buzz around machine learning and IoT, it can be difficult to cut through the noise and understand where the actual value lies. In this week’s #askIoT post, I’ll explain how machine learning can be valuable for IoT, when it’s appropriate to use, and some applications and use cases currently out in the world today.

With rumors swirling about Apple’s plans in Augmented Reality (AR), as well as recent teeth gnashing about the state of Magic Leap, it’s beyond time for the AR community to have a real discussion about what it’s going to take for Augmented Reality to become the primary computing environment worldwide.

The question is, what will it take for an AR system to replace the smartphone and the desktop as the ubiquitous computing device for everyday use?

I’ve spoken in the past about what I call the “AR stack.” That is the set of technologies that are necessary to get to wide functionality and applicability for the average consumer. However I wanted expand on that and build a very basic reference that identifies key concepts necessary for AR, that people can easily cite when trying to understand this topic.

The joint many-task model tackles multiple NLP tasks with a single architecture. Tasks are layered such that subsequent and previous tasks benefit from training of the closely-related tasks. Though applied to specific NLP objectives, the proposed model introduces a powerful concept for future research.

As a follow-up of my previous post on reliability diagrams, I have worked jointly with Alexandre Gramfort, Mathieu Blondel and Balazs Kegl (with reviews by the whole team, in particular Olivier Grisel) on adding probability calibration and reliability diagrams to scikit-learn. Those have been added in the recent 0.16 release of scikit-learn as CalibratedClassifierCV and calibration_curve.

This post contains an interactive version of the documentation in the form of an IPython notebook; parts of the text/code are thus due to my coauthors.

Parallel Forall is a GPU Computing developer blog, focused on providing detailed technical information on a variety of massively parallel programming topics, including CUDA C/C++, OpenACC, GPU-accelerated libraries, Machine Learning, GPU programming techniques, and much more. At Parallel Forall you will find useful information about productive, high-performance programming techniques for the latest GPU technology. Readers have opportunities for discussion with leading experts in GPU computing, and topics featured in Parallel Forall are of interest to programmers of all levels, and from a variety of disciplines. Parallel Forall is curated by Mark Harris, Chief Technologist for GPU Computing Software at NVIDIA.

This is the blog of Fabian “ryg” Giesen. I work at RAD Game Tools in Kirkland/WA as a programmer. I’m also active in the demoscene group Farbrausch and have written some useful tools and other pieces of code, most of which are available on my homepage.

This is the index page for a series of blog posts I’m currently writing about the D3D/OpenGL graphics pipelines as actually implemented by GPUs. A lot of this is well known among graphics programmers, and there’s tons of papers on various bits and pieces of it, but one bit I’ve been annoyed with is that while there’s both broad overviews and very detailed information on individual components, there’s not much in between, and what little there is is mostly out of date.

This series is intended for graphics programmers that know a modern 3D API (at least OpenGL 2.0+ or D3D9+) well and want to know how it all looks under the hood. It’s not a description of the graphics pipeline for novices; if you haven’t used a 3D API, most if not all of this will be completely useless to you. I’m also assuming a working understanding of contemporary hardware design – you should at the very least know what registers, FIFOs, caches and pipelines are, and understand how they work. Finally, you need a working understanding of at least basic parallel programming mechanisms. A GPU is a massively parallel computer, there’s no way around it.

Our World in Data (OWID) is an online publication that shows how living conditions are changing. The aim is to give a global overview and to show changes over the very long run, so that we can see where we are coming from and where we are today.

Our World in Data communicates this empirical knowledge in two ways:
1. through data visualizations – charts and maps.
2. by presenting the academic research on global development that explains what drives the changes that we see and what the consequences of these changes are.

The publication is produced at the University of Oxford and currently created by a team of three: The economist Esteban Ortiz Ospina, the web developer Jaiden Mispy, and the founder of the publication Max Roser.

I was talking the other day with a former student at UW, Sarah Rich, who’s done degrees in both math and CS and then went off to Twitter. I asked her: so what would you say to a math Ph.D. student who was wondering whether they would like being a data scientist in the tech industry? How would you know whether you might find that kind of work enjoyable? And if you did decide to pursue it, what’s the strategy for making yourself a good job candidate?

Sarah exceeded my expectations by miles and wrote the following extremely informative and thorough tip sheet, which she’s given me permission to share.

We’ve always thought that the world’s largest encyclopedia should have a world-class library. Through the Wikipedia Library program, the encyclopedia’s editors have free access to a collection of over 80,000 unique periodicals, like journals, magazines, newspapers, newsletters, pamphlets, and series, in addition to an untallyable number of books. This access has been facilitated by over 60 partners, including many of the world’s leading publishers and aggregators.

The library of resources available to Wikipedians continues to grow, allowing these editors to use the best sources available to improve Wikipedia.