As the use of machine learning continues to expand, the importance of ensuring its safety cannot be overstated. A key concern in this regard is the ability to identify whether a given sample is from the training distribution, or is an "Out-Of-Distribution" (OOD) sample. In addition, adversaries can manipulate OOD samples in ways that lead a classifier to make a confident prediction. In this study, we present a novel approach for certifying the robustness of OOD detection within a $\ell_2$-norm around the input, regardless of network architecture and without the need for specific components or additional training. Further, we improve current techniques for detecting adversarial attacks on OOD samples, while providing high levels of certified and adversarial robustness on in-distribution samples. The average of all OOD detection metrics on CIFAR10/100 shows an increase of $\sim 13 \% / 5\%$ relative to previous approaches.

Since then, I've had the chance to use Docker more in my work and have picked up some new tricks. This was long overdue, but I've finally collected my updated learnings in this post. Recently, I encountered an article titled ROS Docker; 6 reasons why they are not a good fit, and I largely agree with it. However, the reality is that it's still quite difficult to ensure a reproducible ROS environment for people who haven't spent years fighting the ROS learning curve and are adept at debugging dependency and/or build errors… so Docker is still very much a crutch that we fall back on to get working demos (and sometimes products!) If the article above hasn't completely discouraged you from embarking on this Docker adventure, please enjoy reading.

March 22nd, 2012 is the day it all began. That's the day we officially incorporated the Open Source Robotics Foundation, the origin of what we now call Open Robotics. The prospect of starting a company is both scary and exciting; but starting an open-source company in a niche as specialized as robotics, now that is terrifying and exhilarating, if not a little unorthodox. All we had was a dream, some open-source code, and some very smart friends, a whole lot of them. We also had the wind at our backs.

Windows 10 version 2004 and on contains something called WSL 2.0, which is an actual lightweight Virtual Machine (VM). Before we talk about what the heck WSL 2.0 is, however, it would be nice to understand a bit about what WSL 1.0 is. WSL 1.0 is Window's answer to something called Cygwin. Cygwin was popular on older versions of Windows for developers who needed or wanted Linux-related tools and a bash shell (a GUI could be added after jumping through a few hoops, though, but unless you had nice hardware the GUI would've been pretty slow -- my experience a bit buggy, too). The suits and ties at Microsoft haven't given the go ahead for WSL 2.0 to support the Linux GUI …OR… the devs are running behind but the good folks at Kali solved it.

The Port of Rotterdam's blockchain subsidiary, Blocklab, has been trialing a decentralized electricity trading system to help lower costs and optimize the use of renewables on its microgrid. The system, called Distro, has been jointly developed by Blocklab and S&P Global Platts and has been operational as a trial for two months. Distro uses blockchain technology, smart contracts and artificial intelligence to support the decentralized, high-frequency trading of renewable energy by commercial consumers looking to optimize and manage their energy use. It matches demand with the intermittent power generated from different sources, specifically solar and battery storage. Each market participant is allocated an AI energy-trading agent that learns their behavior, choices and needs and provides them with energy at the optimal price.

The Port of Rotterdam's blockchain subsidiary, Blocklab, has been trialing a decentralized electricity trading system to help lower costs and optimize the use of renewables on its microgrid. The system, called Distro, has been jointly developed by Blocklab and S&P Global Platts, and has been operational as a trial for two months. Distro uses blockchain technology, smart contracts, and artificial intelligence to support the decentralized and high frequency trading of renewable energy by commercial consumers looking to optimize and manage their energy use. It matches demand with the intermittent power generated from different sources, specifically solar and battery storage. Each market participant is allocated an AI energy trading agent that learns their behavior, choices, and needs and provides them with energy at the optimal price.

For 1 or 2 years I searched for a linux distro for AI. It seemed to me that such a distro should exist, because artificial intelligence is a hot topic, and probably many people, like me, would love to dive in and tinker on a ready-made platform. After all, there are many Linux distros tailored to niche interests, why not AI? But I didn't find anything like that. Having once before created a linux distro I decided that maybe I would make another.