The world has changed, and the protagonists of this change are small, agile, laser-focused startups that quickly innovate and deploy products in the market at a fraction of the cost of what a large enterprise can do. The ‘democratization of high tech innovation’ is a pervasive trend in both commercial and defense applications, and the trend has been picked up by the US Department Of Defense (DoD).
May 13, 2016|
February 17, 2016|
The U.S. Patent Office issued to Neurala, Inc. a new patent (US 9,189,828) that extends the US patent 8,648,867 (Graphic Processor Based Accelerator System and Method) into the domain of real-time control of autonomous machines, such as self-driving cars and drones.
While the prior patent covered hardware and software “controllers” that handle most of the primitive operations needed to set up and control Deep Networks and Neural Networks on a GPU, the current patent extends the prior one by providing specific indication on how to this system could control real-time operating machines, such as drones and self-driving cars.
Why is this important?
February 6, 2016|
When designing robotic platforms, the choice of which sensors to employ is a key area that often determines the Go/No-Go for a final product. This is because the cost of sensors is a huge component of the total cost of robots, and the main challenge in front of effective commercialization of consumer robotic platforms and applications. This is true at all levels: from inexpensive consumer robots, to drones (which had payload issues, among others!), all the way to self-driving cars. Read the rest of this entry »
January 25, 2016|
According to ABI Research, a technology market intelligence firm in Oyster Bay, New York, by 2025, sales of drones to the consumer market are expected to exceed 90 million units and generate $4.6 billion in revenue!
ABI said that in 2014, there were 4.9 million drone sales, but the predicted increase in drone sales will create a compound annual growth rate of 30.4 percent over the next 10 years.
Full article available in this link!
January 20, 2016|
A slightly different post ..... away from robot and GPU! Today, I want to point to a new article titled Resonant Cholinergic Dynamics in Cognitive and Motor Decision-Making: Attention, Category Learning, and Choice in Neocortex, Superior Colliculus, and Optic Tectum was published on Frontiers in Neuroscience (20 January 2016 | http://dx.doi.org/10.3389/fnins.2015.00501). The article, directly derived from my PhD research on laminar mechanisms of memory and perception in the cerebral cortex, was written in collaboration with Jesse Palma. This article, freely available online, elaborates on links between ... a cup of coffee (cholinergic modulation!!!!), attention, learning, and decision making within several brain systems, both cognitive and motor.
Selecting relevant sensory information while interacting with a changing environment is a key feature not only in robotsof animal intelligence. This selection is necessary to direct limited sensory, cognitive, and motor resources toward the important stimuli in the environment, and to choose a set of motor commands that correspond to behavioral goals. The present article proposes how cholinergic modulation of cognitive and sensory-motor circuits may realize such selectivity in a task-sensitive way. Understanding how animals can solve this, can help build more intelligent machines.Categories: Neurobiology
December 31, 2015|
There are three main ingredients needed for intelligent robots to be ubiquitous, smart, and useful. I like to call these three ingredients Mind, Brain, and Body. Let's look at how these three enabling technology have evolved, and why the time is now for the emergent of intelligent machines. Read the rest of this entry »
December 21, 2015|
Anniversaries are nice, especially in December! So here is one to share with you all: it has been 10 years since Anatoli, Heather, and Max (the 3 co-founders of Neurala) begun tinkering with GPUs and Deep Learning, setting the foundations of Neurala while getting their Ph.D at Boston University. This post is a re-publication of an earlier post based on our work on AMD GPUs. Funny to say, much of the techniques, which are now in an awarded patent, are still the backbones of Neurala's tech. But let's get back to the past!
By Anatoli Gorchetchnikov, Heather Ames, Massimiliano Versace
The last post on GPU made me think of a project Anatoli Gorchetchnikov, Heather Ames and myself embarked on in 2006 when we got really interested in general purpose computing on graphic processing cards. At the time, there was no CUDA or OpenGL available: programming GPUs was really tough. But we tried, with some very good results, to port some of the models we used on GPUs. Here is how we did it. Read the rest of this entry »
November 24, 2015|
A few years ago, Heather’s, Anatoli, and myself started Neurala with one goal: bring the results and insights of our Ph.D. work on brain-inspired computing into everyday technology. We wanted this technology to change the way society uses and benefits from machines: rather than each device needing 1 human brain to work, we wanted it to have "its own brains" at the service of the human owner. Simple goal, not so simple technology, but designed from the ground-up to be a technology that would be helpful to humanity. But sometimes people like to think differently.
Read the rest of this entry »Categories: Robotics
November 17, 2015|
One of the pillars of the recent success (almost viral) of Deep Networks, a subspecies of the bigger class called Neural Networks, is that their execution and training methods are highly conducive to parallelism. The term GPGPU is often use to refer to the backbone of the revolution: General-Purpose computation on Graphics Processing Units. GPUs, chips whose main technological push comes from huge revenues from the gaming market, and more recently are finding their ways into mobile devices, are in reality high-performance many-core processors that can be used to accelerate a wide range of applications, going from physics, to chemistry, to computer vision, to neuroscience. And, of course, Deep Networks.
Read the rest of this entry »