- How Transferable Are Features in Deep Neural Networks? — (answer: “very”). A final surprising result is that initializing a network with transferred features from almost any number of layers can produce a boost to generalization that lingers even after fine-tuning to the target dataset. (via Pete Warden)
- Introducing Atlas: Netflix’s Primary Telemetry Platform — nice solution to the problems that many have, at a scale that few have.
- The Many Facades of DRM (PDF) — Modular software systems are designed to be broken into independent pieces. Each piece has a clear boundary and well-defined interface for ‘hooking’ into other pieces. Progress in most technologies accelerates once systems have achieved this state. But clear boundaries and well-defined interfaces also make a technology easier to attack, break, and reverse-engineer. Well-designed DRMs have very fuzzy boundaries and are designed to have very non-standard interfaces. The examples of the uglified DRM code are inspiring.
- DPDK — a set of libraries and drivers for fast packet processing […] to: receive and send packets within the minimum number of CPU cycles (usually less than 80 cycles); develop fast packet capture algorithms (tcpdump-like); run third-party fast path stacks.
"deep learning" entries
The history of computing has been a constant pendulum — that pendulum is now swinging back toward distribution.
The trifecta of cheap sensors, fast networks, and distributing computing are changing how we work with data. But making sense of all that data takes help, which is arriving in the form of machine learning. Here’s one view of how that might play out.
Clouds, edges, fog, and the pendulum of distributed computingThe history of computing has been a constant pendulum, swinging between centralization and distribution.
The first computers filled rooms, and operators were physically within them, switching toggles and turning wheels. Then came mainframes, which were centralized, with dumb terminals.
As the cost of computing dropped and the applications became more democratized, user interfaces mattered more. The smarter clients at the edge became the first personal computers; many broke free of the network entirely. The client got the glory; the server merely handled queries.
Once the web arrived, we centralized again. LAMP (Linux, Apache, MySQL, PHP) buried deep inside data centers, with the computer at the other end of the connection relegated to little more than a smart terminal rendering HTML. Load-balancers sprayed traffic across thousands of cheap machines. Eventually, the web turned from static sites to complex software as a service (SaaS) applications.
Then the pendulum swung back to the edge, and the clients got smart again. First with AJAX, Java, and Flash; then in the form of mobile apps, where the smartphone or tablet did most of the hard work and the back end was a communications channel for reporting the results of local action. Read more…
Step-by-step instruction on training your own neural network.
When I first became interested in using deep learning for computer vision I found it hard to get started. There were only a couple of open source projects available, they had little documentation, were very experimental, and relied on a lot of tricky-to-install dependencies. A lot of new projects have appeared since, but they’re still aimed at vision researchers, so you’ll still hit a lot of the same obstacles if you’re approaching them from outside the field.
In this article — and the accompanying webcast — I’m going to show you how to run a pre-built network, and then take you through the steps of training your own. I’ve listed the steps I followed to set up everything toward the end of the article, but because the process is so involved, I recommend you download a Vagrant virtual machine that I’ve pre-loaded with everything you need. This VM lets us skip over all the installation headaches and focus on building and running the neural networks. Read more…
Announcing a new series delving into deep learning and the inner workings of neural networks.
Editor’s note: this post is part of our Intelligence Matters investigation.
When I first ran across the results in the Kaggle image-recognition competitions, I didn’t believe them. I’ve spent years working with machine vision, and the reported accuracy on tricky tasks like distinguishing dogs from cats was beyond anything I’d seen, or imagined I’d see anytime soon. To understand more, I reached out to one of the competitors, Daniel Nouri, and he demonstrated how he used the Decaf open-source project to do so well. Even better, he showed me how he was quickly able to apply it to a whole bunch of other image-recognition problems we had at Jetpac, and produce much better results than my conventional methods.
I’ve never encountered such a big improvement from a technique that was largely unheard of just a couple of years before, so I became obsessed with understanding more. To be able to use it commercially across hundreds of millions of photos, I built my own specialized library to efficiently run prediction on clusters of low-end machines and embedded devices, and I also spent months learning the dark arts of training neural networks. Now I’m keen to share some of what I’ve found, so if you’re curious about what on earth deep learning is, and how it might help you, I’ll be covering the basics in a series of blog posts here on Radar, and in a short upcoming ebook. Read more…