Click here to download the abstracts of poster presentations.
As machines become ‘smarter’, more autonomous and ubiquitous, how does this impact human creativity, or the role of the artist? In this talk, I’ll briefly cover some of my own investigations in this field with examples spanning research, sketches and artworks; particularly within the context of recent developments in machine (and deep) learning. These include i) explorations in real-time, interactive computational systems to augment artistic expression; ii) semi-autonomous, learning systems with continuous, meaningful human control, for collaborative co-creation; iii) thinking about machine learning algorithms (and computational perspectives in general) as a means to reflect on ourselves and how we make sense of the world; what (and who) we choose to value and why; our own self-affirming cognitive biases and prejudices; our inability to see the world from others’ points of view and empathise with those that we disagree with; and the resulting social and political polarisation and gaping wounds in our societies. In short, I’ll be oscillating around the blurry lines between Waves, and Gods.
“The journey for each of us begins here. We’re going to explore the cosmos in a ship of the imagination, unfettered by ordinary limits on speed and size, drawn by the music of cosmic harmonies. It can take us anywhere in space and time. Perfect as a snowflake, organic as a dandelion seed, it will carry us to worlds of dreams and worlds of facts. Come with me.” – Carl Sagan.
Introduction to Machine Learning
Machine learning (ML) techniques are artificial intelligence (AI) methods, which are utilized in a large variety of application fields. ML methods either explicitly or implicitly construct a statistical/mathematical model of the data based on a sample dataset called ‘training data’. Based on the underlying statistical/mathematical model ML methods can perform predictions or make decisions without being explicitly programmed to do so. As described earlier ML techniques are utilized in a wide range of application like spam detection in email accounts, computer vision, data analysis of spectroscopic data and diagnostic tasks. Deep learning is a special form of ML and these methods are utilized extensively in the last decade. In this contribution an introduction to ML methods including deep learning will be given and especially medical application examples will be presented. Finally the impact of the techniques on society and science will be briefly discussed.
AI and Growth of Knowledge
1- A simplified summary of AI theory (statistical theory of learning) explained in terms of 2D shapes, 2- explicit vs tacit knowledge and their relation in the context of endogenous economic growth theory of Paul Roemer, 3- History of explicit vs tacit knowledge :episteme vs techne in the context of ancient Greek cosmology attributed to Aristoteles 4- A sudden emergence and growth of terrestial natural sciences after Copernicus and Newton 5- Conclusion : relating historical explosion of scientific knowledge to concepts of AI.
A Conversing Intelligent Patient Monitoring System
Demographics of societies have been changing slowly yet steadily. Especially in developed societies, there is a pressing and eminent threat to the sustainability of social security systems owing to the unfortunate increasing trend in elderly-to-young population ratios. With increasing elderly population and accompanying elongation of lifespan, the financial burden of health care expenses of elderly people has been frighteningly increasing.
E3Nurse is one feasible, working, and affordable solution relying on intelligent processing of many vitals collected via a wearable device from an elderly patient at home. E3Nurse, not disturbing or intruding the elderly’s comfortable daily life, checks the vital regularly without stealing the patient’s attention. E3Nurse’s fuzzy intelligent system, upon detecting an anomaly, either makes an emergency call or converses with the patient to evaluate the situation.
Such a standalone system, also keeping health carers and the relatives informed of the patient’s status, is a promising approach to cut down the expenses of elderly care as well as reduce the human effort. With such measurable impacts, E3Nurse can be used for reducing the adverse effects of increase in elderly population and lifespan.