Peter Voss Seeks The Third Wave of AI with Aigo

Tuesday, April 10, 2018

Peter Voss Seeks The Third Wave of AI with Aigo

AGI researcher, futurist and entrepreneur Peter Voss wants to move beyond traditional hand-coded programming and neural networks with the formation of his new startup, Aigo. The company is building intelligent systems that just might overcome the limitations of other AI architectures.

Imagine a super-smart Alexa/ Siri/ Google Assistant-like app or device that actually remembers what you told it previously; can hold lengthy conversations; can interact with other assistants, devices, and apps; and that can learn new facts, preferences and skills via English conversation. 
Now imagine that this assistant isn’t just available in one domain, but can assist you using complex software, allow you to control and monitor your home, car, and other devices, in addition to handling a wide range of personal and business requests.

That is the promise of a new artificial intelligence start-up, Aigo.AI, founded by Peter Voss Voss wants us to imagine that this personal AI not only gets smarter as it learns from us, but also as it interacts with other people (and AIs) in your network. This new AI instantly acquires new knowledge, skills, and capabilities via ongoing development.

Peter Voss
For Voss, the pursuit of artificial general intelligence or AGI is a long-held goal. He even claims to be one of the coiners of the term.

In his long career, Voss has designed novel electronic circuits, developed a 4GL DB and several ERP systems. He has spent the last few years deeply involved in AGI research.

Voss, who is well-known in Singularity circles, also writes about ethics, freewill, artificial minds and futurism, and wants to live a really long time.

"I am now focused on commercializing the second generation of our ‘Conversational AI’ technology (called 'Aigo') that is more like ‘Her’ and less like ‘Dory’," he writes on his LinkedIn profile.

Related articles
For Voss, Aigo is seeking to develop AI systems based on the 'Third Wave' of development. The first of these waves was, based on traditional programming or, GOFAI (Good Old-Fashioned AI). Alan Turing himself would be familiar with this route. This path got us great chess playing systems and tax preparation systems, but these hand-crafted systems could not learn or deal with abstract data well. Such systems do not handle the real world very well.

The second was bases on deep learning and the use of neural networks. The progress we see in self-driving cars, voice assistants and more would not have been possible without this technology. In the last few years, deep learning has been a revolution in AI, with literally tens of billions of dollars being invested to further develop and exploit the technology.

These systems are excellent and learning and categorizing based on statistical methods. These systems are not great at general knowledge handling, or reasoning. Furthermore, second wave systems rely on very large data sets to perform their analysis.

Three waves of artificial intelligence

Voss calls the Third wave 'cognitive architecture.' "The First and Second Waves gave us a glimpse of what AI can potentially do for us; we’re looking to the Third Wave to more fully realize the potential of AI," he writes.

Cognitive architectures approaches have a long history in AI. At the heart, it is the idea of having a core ‘cognitive engine’ that supports all the basic functionality required for general intelligence, according to Voss. It is a uniform, highly-integrated system of support structures and functions that, working together, produce all the cognitive abilities required for general intelligence. These include short- and long-term memory, pattern matching, prediction, prioritization, reasoning, planning, and many others.

Similar to the way that deep learning neural nets had been around for several decades before exploding into usefulness and prominence, so have cognitive architectures been lurking in the AI background for a long time. "I believe that given the right design, cognitive architectures can provide the path to AGI," states Voss.

This is the Third Wave. A presentation last year by John Launchbury, the Director of DARPA's Information Innovation Office described the development path. “In the third wave, the AI systems themselves will construct models that will explain how the world works," Launchbury said.  For Launchbury, the second wave neural nets can be simplified as "spreadsheets on steroids," and this fact highlights the limitations of such systems, and invites the development of another approach.

According to Voss, applying a cognitive architecture is not, by itself, adequate for achieving real intelligence. To be truly effective, the cognitive components also need to be deeply integrated and tightly coupled. "Knowledge representation must fully integrate knowledge and skills, which in turn need to integrate with short-term memory, context, reasoning, " he claimed in an interview on Towards Data Science, "This fact is borne out both by my theoretical model and by practical implementation experience."

Third Wave of artificial intelligence

This approach inherently seeks to account for all relevant requirements of human cognition, including such aspects as knowledge representation, generalized short-and long-term memory, perception, focus, goal management, etc.
The Aigo architecture follows this path. According to the company's website, Aigo’s core technology, leapfrogs current machine learning approaches and implements a sophisticated graph-based cognitive architecture that overcomes many of the inherent limitations of big data/ statistical methods typical of second wave methods.

The company is just emerging from stealth, so there aren't may concrete items to evaluate yet, but Voss has promised demos to be released soon.

By  33rd Square