The rise of Artificial Intelligence (AI), and an increasingly connected society has already, according to the UK's MI5 made it "much harder for spies to hide in the shadows", but now, if Robert Cardillo has his way, so called robo-automation tools will perform 75 percent of the tasks currently done by the new front line of American intelligence spies – the analysts who collect, analyse, and interpret images beamed from drones, satellites, and other feeds around the globe. Cardillo, the director of the National Geospatial-Intelligence Agency, (NGA), announced his push toward "automation" and "AI" at a conference this week in San Antonio. The annual conference, hosted by the United States Geospatial Intelligence Foundation, brings together technologists, soldiers, and intelligence professionals to discuss national security threats, changes in technology, and data collection and processing. AI is on the rise, and last year former President Barack Obama's White House created a Defcon Scale for Cyberattacks, and released a white paper on its potential future impacts in the final months of the administration, and police forces around the world are increasingly using preliminary "pre-crime" technologies to predict when, where and by whom crimes will likely be committed. And all of that is in addition to the likes of companies like Amazon and Netflix who are using machine learning to calculate what movie you will want to watch or which book you may buy.

As technologies such as artificial intelligence (AI) and machine learning (ML) mature, innovation is no longer the sole domain of consumer-facing businesses. With the democratisation of technology, organisations in the business-to-business (B2B) space -- traditionally deemed as slow adopters -- are enthusiastically embracing AI technology. The objective is to offer advanced and responsive products and services to clients. For example, with the help of AI and ML, Otis is developing smart elevators capable of communicating with passengers, building managers, service staff and other building systems. The elevator manufacturer is transforming its service business to incorporate smart, connected technology that delivers proactive, quick and effective diagnostics and repair.

Artificial intelligence (AI) is all the buzz this year. According to CB Insights, as of June 15, this year, more than 200 AI venture financing deals have been completed already totaling $1.5B in dollar volume. If the latter half of the year continues at this pace, 2016 will be a record year. Based on analysis by CB Insights, most of the deals being done are at a series B or C stage, indicating that startups in this space are beginning to see success. AI, as most people now know, has several applications in health technologies, marketing & sales, business analysis and financial services.

I've been putting together a map on Canada's AI ecosystem, which I first revealed last week in my keynote on C2 Montreal's main stage. As promised, I'm publishing that map at the bottom of this post. Given the speed at which the industry is progressing, this map is constantly evolving, so I'll be sharing updates as we add them. If you have an addition to make, drop me a line! UPDATE June 13, 2017: Last week I posted V1 of my Map of the Canadian AI Ecosystem, and since then I've been inundated with additions.

UPDATE June 13, 2017: Last week I posted V1 of our Map of the Canadian AI Ecosystem, and since then I've been inundated with additions. While it's still incomplete, I thought it was important to update the map currently being sent around and reinforce the idea that this ecosystem is in constant flux. Already, our list has mushroomed from about 160 startups to over 550. Goes to show how much is happening just below our attention. I've been putting together a map on Canada's AI ecosystem, which I first revealed last week in my keynote on C2 Montreal's main stage.

One measure of the status of civilization is the complexity of tools used by the society. As societies have progressed, tools and machines used by them have become increasingly complex. Despite their rising complexity the current set of tools and machines still need humans to create and use them and they can only do things what humans have pre-programmed them to do or control them to do. In particular, current set of machines cannot learn and enhance their knowledge. However, a new set of machines are emerging that can learn and they need minimal human intervention to operate.

We have discussed some AI topics in the previous posts, and it should seem now obvious the extraordinary disruptive impact AI had over the past few years. However, what everyone is now thinking of is where AI will be in five years time. I find it useful then to describe a few emerging trends we start seeing today, as well as make few predictions around machine learning future developments. The following proposed list does not want to be either exhaustive or truth-in-stone, but it comes from a series of personal considerations that might be useful when thinking about the impact of AI on our world. Companies like Vicarious or Geometric Intelligence are working toward reducing the data burden needed to train neural networks.

While the growth of deep neural networks has helped propel the field of machine learning to new heights, there's still a long road ahead when it comes to creating artificial intelligence. That's the message from a panel of leading machine learning and AI experts who spoke at the Association for Computing Machinery's Turing Award Celebration conference in San Francisco today. He said that applications using neural nets are essentially faking true intelligence but that their current state allows for interesting development. "Some of these domains where we're faking intelligence with neural nets, we're faking it well enough that you can build a company around it," Jordan said. Those comments come at a time of increased hype for deep learning and artificial intelligence in general, driven by interest from major technology companies like Google, Facebook, Microsoft, and Amazon.

Two driverless shuttles will begin operating at the University of Michigan this fall. The tall, airy, 15-passenger shuttles will carry students and staff in a two-mile loop on campus roads alongside regular traffic. The shuttle will be free and insured by the university. A driverless shuttle carries passengers at the University of Michigan, Wednesday, June 21, 2017, in Ann Arbor, Mich. Two driverless shuttles will begin operating at the university this fall. The tall, airy, 15-passenger shuttles will carry students and staff in a two-mile loop on campus roads alongside regular traffic.

We develop a simple routine unifying the analysis of several important recently-developed stochastic optimization methods including SAGA, Finito, and stochastic dual coordinate ascent (SDCA). First, we show an intrinsic connection between stochastic optimization methods and dynamic jump systems, and propose a general jump system model for stochastic optimization methods. Our proposed model recovers SAGA, SDCA, Finito, and SAG as special cases. Then we combine jump system theory with several simple quadratic inequalities to derive sufficient conditions for convergence rate certifications of the proposed jump system model under various assumptions (with or without individual convexity, etc). The derived conditions are linear matrix inequalities (LMIs) whose sizes roughly scale with the size of the training set. We make use of the symmetry in the stochastic optimization methods and reduce these LMIs to some equivalent small LMIs whose sizes are at most 3 by 3. We solve these small LMIs to provide analytical proofs of new convergence rates for SAGA, Finito and SDCA (with or without individual convexity). We also explain why our proposed LMI fails in analyzing SAG. We reveal a key difference between SAG and other methods, and briefly discuss how to extend our LMI analysis for SAG. An advantage of our approach is that the proposed analysis can be automated for a large class of stochastic methods under various assumptions (with or without individual convexity, etc).