ThinkTV is proud to announce the formation of an independent laboratory to carry out cutting-edge research into the performance of TV advertising, in partnership with leading international media academic Professor Karen Nelson-Field (pictured above) and Media Intelligence Co. The laboratory's forthcoming two-year research program will help advertisers and media agencies get the best out of TV by providing robust evidence and greater clarity about how multiplatform TV advertising delivers business results. The ThinkTV Smart Lab is directed by Karen Nelson-Field, Professor of Media Innovation at the University of Adelaide and CEO of research joint venture, Media Intelligence Co. (MIC). The purpose-built facility will examine TV's impact on brand and advertiser performance and will use artificial intelligence technologies to remove human error and bias. It is funded by ThinkTV but remains independent to ensure research rigour and credibility.

THIS is a timely and provocative anthology on a theme that has fascinated scientists, philosophers and SF writers for decades. The "singularity" occurs when artificial intelligence (AI) exceeds that of humans and AIs design new technologies beyond our understanding. Its realisation would bring massive and unpredictable changes to civilisation and the environment. The book's symposium structure has a "target" paper by philosopher David J Chalmers which elicits a range of articles in response and it concludes with a response from the same author, while the editor's introduction establishes the history of the singularity concept, highlights the interdisciplinary debates it generates and provides an accessible way into the language and logical arguments employed in later, more challenging pieces. Chalmers argues that human-level AI is likely to be created in a century or so, unless prevented by disaster, legislation or direct action. If AI is created, AI -- representing greater than human intelligence -- is likely to follow within decades and, a few years after, the world would see AI systems with much greater than human intelligence.

Honeypot design and deployment is a tradeoff between realism and simplicity; this tradeoff can be characterized as the difference between high and low interaction honeypots. A realistic design could use an actual operating system instrumented to detect and capture intruders (known as a high interaction honeypot). However, the detection would be greatly complicated, because it is difficult to distinguish between normal traffic on the system and the attacker's. It is a low signal to noise detection problem due to the complexity of modern operating systems running hundreds of threads generating large volumes of traffic with complex signatures. A honeypot that is designed only to superficially mimic an OS (low interaction honeypot) can easily detect the attacker's actions, since there is no background noise.

I've been trying to think of a way to describe how big Machine Learning is, and I think I finally have a decent one: Because Statistics is the primary mechanism we've had for decades to learn from data. That's what Machine Learning is, except far more powerful. Most importantly, machine learning can…well, learn. It improves as it gets more data. More wisdom potentially gets extracted when you apply Statistics to more (and better) data, but the analysis itself doesn't improve with better data.

Google has tweaked its "deep learning" AI to use an external memory bank. It's an attempt to replicate the way human brains use short term memory to simplify reasoning. The company demonstrated the approach by having the system teach itself the London Underground (subway) map and figure out the quickest route between stops. It's a simple task to humans, but the process – which involves comparing multiple branching options with 270 stops over 11 lines – is exactly the type of problem that poses a challenge to artificial intelligence. Because the system was allowed temporary access to stored memory, it was able to more effectively process and categorize the possible routes without having to start from scratch each time. That's similar to how a human brain could use short term memory to filter down all the possible routes by ruling out every one that involves travelling in a particular direction from a specified stop, repeating the process until only the optimum answer remained.

We think the next era will belong to "the conversational layer" -- both text- and voice-driven -- that will use chat, messaging, or natural language interfaces to interact with people, brands, services, and bots. This interaction will occur at the exact time the user demands a product or service, and in the exact terms she thinks of that product or service, in the language and communication methods she typically uses (intent, words, shortcuts, emojis, etc.). In order to reach user conversations today, brands will need to decide which platforms to target and build on. Many of the tools that are provided by the messaging and bot platform providers are from the open-source space, and companies can perform low-cost experiments with a reduced set of users to learn more about conversational interactions and use cases that yield the desired results.

Originally posted here, where you can see all the graphics. There has been much in the news lately about the next wave of MT technology driven by a technology called deep learning and neural nets (DNN). I will attempt to provide a brief layman's overview about what this is, even though I am barely qualified to do this (but if Trump can run for POTUS then surely my trying to do this is less of a stretch). Please feel free to correct me if I have inadvertently made errors here. Traditional AI methods of language understanding depended on embedding rules of language into a system, but in the Google SmartReply project, as with all modern machine learning, the system was fed enough data to learn on its own, just as a child would.

Backpropagation is a common method for training a neural network. There is no shortage of papers online that attempt to explain how backpropagation works, but few that include an example with actual numbers. This post is my attempt to explain how it works with a concrete example that folks can compare their own calculations to in order to ensure they understand backpropagation correctly. If this kind of thing interests you, you should sign up for my newsletter where I post about AI-related projects that I'm working on. You can play around with a Python script that I wrote that implements the backpropagation algorithm in this Github repo.