Tag: algorithm

Deep learning for the kids with Lobe

Friday, 4 May, 2018 0 Comments

“Lobe is an easy-to-use visual tool that lets you build custom deep learning models, quickly train them, and ship them directly in your app without writing any code. Start by dragging in a folder of training examples from your desktop. Lobe automatically builds you a custom deep learning model and begins training. When you’re done, you can export a trained model and ship it directly in your app.”

Just a decade ago, the algorithmic complexity Lobe offers sounded like science fiction. Now, it is freely available for anyone with a computer and Lobe founder Mike Matas says the visual interface is so user-friendly that children can understand it. So, c’mon young and old, it’s time “to build custom deep learning models, quickly train them, and ship them directly in your app without writing code.”

Note: If you want to talk the deep learning talk and walk the deep learning walk, there’s no avoiding Python and cloud, calculus and linear algebra, however. “How to learn Deep Learning in 6 months” provides some useful pointers.

The algorithm will read your story today, Homer

Wednesday, 15 February, 2017 0 Comments

Modern mythology is created in film studios. There, the stories the global village has come to depend on for its entertainment and inspiration are turned into drama, comedy, action, science fiction, fantasy, thriller, horror, crime, noir, epic, western, war, romance, musicals, blockbusters…

As in the days of Homer, the secret of success is to tell the tale with emotion and imagery that the audience cannot forget. Easier said than done, of course, but if you’ve got $43,000 the American Film Institute Conservatory in Hollywood Hills will take it and, in return, you’ll get some of that old storytelling magic. Blurb: “Tomorrow’s storytellers are placed in a hands-on, production-based environment and are trained by a group of dedicated working professionals from the film and television communities.” In the end, however, it all comes back to Homer: “The AFI Conservatory doesn’t believe there is a formula for making films. Fellows are instead encouraged to find and develop a unique voice as they are trained in the art of storytelling.”

This is all very touching but it doesn’t mention the Epagogix algorithm. Epagogix is a privately-held UK company that “brings together expertise in risk, finance, artificial intelligence and film analysis to create innovative tools and solutions for the hard decisions that senior company directors need to make.” Yes, that’s right, “film analysis.” So, how does it work? Here goes:

Epagogix works confidentially with the senior management of major film studios, large independents and other media companies, assisting with the selection and development of scripts by identifying likely successes and probable ‘Turkeys’; helping to quantify a script/project’s commercial success; and advising on enhancements to the Box office/audience share potential.

Epagogix’s approach helps management of this most critical financial risk by delivering accurate predictive analysis of the Box Office value of individual film scripts, and by identifying and quantifying how and where to improve their commercial value. If requested, Epagogix sensitively bridges the gap between the financial and creative aspects of film production by providing quantified insights and advice to those responsible for script development.

Note the advice “to those responsible for script development.” That’s you, Homer.


Google Translate goes AI

Wednesday, 28 September, 2016 0 Comments

From now on, Google Translate will rely more on AI (artificial intelligence) when it translates languages. Alphabet, the parent company, claims that its brand new Google Neural Machine Translation system will reduce errors by 80 percent compared to its current method.

Google Translate Until today, Google has used what is called “phrase-based translation,” which is standard for the industry. With this method, a hand-coded algorithm breaks down a sentence into words or phrases and tries to match them a vast dictionary. The new system will use that same large dictionary to train two neural networks, one of which will deconstruct the original sentence to figure out what it means, while the other generates text in the output language.

Because AI algorithms don’t rely on human logic, they can often find better ways to do the job compared to the hand-coded algorithms, say the engineers. And as the network learns how to translate, no longer spending time dividing sentences into words or phrases, it discards the rules that humans thought were best and concentrates fully on the outcome. Such is the nature of AI. As Alan Turing wrote in 1950: “I believe that at the end of the century the use of words and general educated opinion will have altered so much that one will be able to speak of machines thinking without expecting to be contradicted.” (Computing machinery and intelligence). We’re getting there.

Google is releasing its new translation system for Mandarin Chinese first, and then adding new languages over coming months.

It’s different this time

Tuesday, 3 February, 2015 0 Comments

Since the Industrial Revolution, there’s been an almost insatiable demand for labour, despite the relentless advance of technology. So why should it be any different this time. Surely, the cloud will create millions of jobs and the app industry will generate global employment? Well, yes, maybe. But let’s consider this: It took the United States some 200 years to change from an agricultural economy, where 90 percent of the people worked on farms, to the current situation, where the number is nearer two percent. The robotics/AI revolution is happening faster than its industrial and digital predecessors — and it will present an even bigger challenge.

Technologies such as the self-driving car will be dramatically disruptive, but over a much shorter time-frame. There are millions of truck drivers working today. What will happen if self-driving vehicles put them out of a job in a matter of years? Algorithms are getting better at translating and writing — jobs that once required humans. So what will we do for work? That is the question being posed by the MIT academics Erik Brynjolfsson and Andrew McAfee, who say that we’re entering a “Second Machine Age,” where the increasing rate of change driven by information technologies could leave swathes of medium-and-low skilled workers in the slow lane. On the upside, the human ability to innovate offers grounds for hope. They say.

The algorithm and the Three Lions

Tuesday, 10 June, 2014 0 Comments

FiveThirtyEight has launched an interactive thingy that calculates every team’s chances of advancing past the group stage in the World Cup and eventually winning the trophy. The forecasts are based on the Soccer Power Index (SPI), an algorithm Nate Silver developed in conjunction with ESPN. For England, SPI predicts unrosy tournament prospects:

“Betting markets see England, Italy and Uruguay as about equally likely to advance while Costa Rica is in a distant fourth place. SPI, by contrast, has England and Uruguay ahead of Italy and views the group as middling enough that Costa Rica could pull off a huge upset…

…It also might not matter much in the end. England, Italy and Uruguay are the sort of teams that might be able to entertain championship dreams in a World Cup with more parity, but not in one where they would have to overcome Brazil, Argentina, Germany or Spain at some point.”

Group D