Arguably the first company to leverage the technology of neural-symbolic computing for the purpose of personal development, Illuminoo hosted an enjoyable afternoon and evening on June 24 at Bloomingdale Beach in the Netherlands. This second edition of Illuminoo Enlights, featured inspiring talks from world-leading experts on personal AI and human-machine interaction, as well as testimonials from users of some of the Illuminoo agents. It was a great success and brought together researchers and entrepreneurs in an informal and sunny atmosphere to talk about the latest achievements in Personal AI, Neural-Symbolic computing and its effect on businesses and consumers worldwide in health, work, sports and education. Detailed information on the program and speakers can be found here.
We kicked-off with a strawberry smoothie on one of the relaxing sofas when we started with a short overview of what Illuminoo is all about and where we are heading for. As a young startup in personal AI (only 1 year old) our ambitions are of course sky-high and we are very committed in achieving them. We are currently working on intelligent agents in trading, health and sports for consumers and major clients, and expect to have three of these already operational before the end of the year. All based on our so-called reflective AI technology, NOOMI®, which learns to know you personally and runs completely on your own device instead of the cloud. In addition, we announced a new consumer website being launched before the end of the year, where you can order and download your own custom-built personal AI, and that Illuminoo will be presenting again at the WebSummit in Lisbon this year as beta-startup (hope to see you there!).
After the kick-off, Artur d’Avila Garcez, leading scientist in Neural-Symbolic Learning and Reasoning, continued with a session on “Reflective Intelligence for Personal AI”, where he, Tarek R. Besold and Mohammed Shameer Iqbal presented their perspective on the ambitions and latest achievements in combining Human Intelligence with Artificial Intelligence for the benefit of both. As Donald Michie used to refer to the existence of AI and Machine Learning (ML) systems, there are three types: systems that have been programmed to perform a difficult task but cannot improve over time, systems that can improve over time, and systems that, in addition, can help humans improve over time. And despite all the recent success of ML, notably the impressive win of the Google DeepMind AlphaGo neural network over professional Go player Lee Sedol, the vast majority of such ML systems are still of type 2, and so far AlphaGo didn’t help us learn how to play Go any better. So to learn from AI, and to understand it better, humans need to be able to interact with a personal AI beyond the black box, to ask questions via different modalities (e.g. natural language, images and movement) and obtain accurate, comprehensible and logically well-structured answers, and above all to describe and re-use such reflective intelligence in intuitive and practical applications.
The second session, hosted by Jeroen Arendsen, our lead UX designer, was all about collaboration between AI and humans and making this successful. As humans love to play games, game play is an important ingredient in helping the two species learn from each other. But what happens when NOOMI agents meet Homo Ludens? Will Man and Machine also play together? Can they both enjoy it and benefit from it? In three very inspiring (almost philosophical) talks, Jeroen Arendsen, Lars Immerthal and Agali Mert shed some light on how to give AI a personal face that users can interact with; that they can understand, argue and play with, on how you engineer good ‘game rules’ and on how applications can benefit from good game design and applying the metaphor of life to support humans in their everyday life as a kind of false-works. In short, they gave us a glimpse on the future of personal AI and the conception of a new generation of digital species to play with and support us.
Alive and Kicking
In the last session, some of the new upcoming agent species were presented by their early adopters. As described before, Illuminoo is working hard to get these intelligent agents operational before the end of the year, but some of them are already in use and have shown great potential. For example, a Magic Mirror which is deployed at Sherpa and entertains and supports people with mental disabilities (by Sylvia Loos), a smart camera that is redefining the rules in sports analytics and broadcasting (by Martijn Brouns), and an agent that knows all about energy trading at Eneco (by Charel Hakkert). Common trade in all these agents is that they share the same reflective intelligence and playful drive to support people in their everyday lives, whether it is in health-care, sports, work or education. They are becoming alive and kicking, so to speak, in the sense that they help us move in the right direction by showing the way through personal insights, still leaving room to make our own conclusions.
Final thoughts and inspiration
After a fun-filled and inspirational day at the beach, with great people, food, drinks and music (see my kids hanging at the bar :-), I can only say that I enjoyed Illuminoo Enlights 2017 very much, and I hope you did as well!
And if you couldn’t be there or like to know more about Illuminoo or our plans for Illuminoo Enlights 2018, please let us know at email@example.com.
With kind regards,
Leo de Penning
And last but not least, we also donated a total of 250 euro (5 euro per ticket) in order to help the victims of war and hunger in Africa. Although this is a small gesture, we are proud to do so, and we strongly urge you to do the same. It is both the most easiest and generous thing to do (donate at Giro555)!