Workshop on Artificial Intelligence and Smart Materials Systems
On September 13th and 14th, starting at 9.00 a.m., the Department of Excellence in Robotics & AI presents the 'Workshop on Artificial Intelligence and Smart Materials Systems'. The workshop will take place in Aula Magna (Scuola Superiore Sant'Anna).
The programme includes two keynotes: the first, by Fabio Roli (University of Genoa), is entitled 'From known to unknown unknowns in AI: Historical and Technical Issues'; the second, by Martin Kaltenbrunner (University of Linz), is entitled 'Sustainable Materials and Design Approaches for Soft Electronics and Robotics'.
Full details in the programme and the attached posters.
The workshop organisers recommend in-person participation. However, it is possible to request a remote connection by contacting Dr Mariangela Barbarito (mariangela.barbarito@santannapisa.it; tel. 050/8833169). Dr. Barbarito can be contacted by Monday, Sept. 12, 1 p.m. to report any intolerances or specific dietary needs.
Abstract keynote Fabio Roli
AI has been originally developed for closed-world, and noise-free, problems where the possible states of natures and actions that a rationale agent could implement were perfectly known. One could argue that, at that time, AI dealt with known knowns. Since the 1980s, when machine learning became an experimental science, AI researchers started to tackle pattern recognition problems with noisy data, using probability theory to model uncertainty and decision theory to minimize the risk of wrong actions. This was the era of known unknowns, characterized by the rise of benchmark data sets, larger and larger year after year, and the belief that real world problems can be solved collecting enough training data. However, recent results have shown that available data sets have often a limited utility when used to train pattern recognition algorithms that will be deployed in the real world. The reason is that modern machine learning has often to face with unknown unknowns. When learning systems are deployed in adversarial environments in the open world, they can misclassify (with high-confidence) never-before-seen inputs that are largely different from known training data. Unknown unknowns are the real threat in many security problems (e.g., zero-day attacks in computer security). In this talk, I give a historical and technical overview of the evolution of AI and machine learning for pattern recognition, and discuss how this evolution can be regarded as a transition from known knowns to unknown unknowns, and the key role that adversarial machine learning plays to make AI safer.
Abstract keynote Martin Kaltenbrunner
Modern societies rely on a wide range of electronic and robotic systems, with emerging stretchable and soft form factors enabling an ever more intimate integration of the digital and biological spheres. These advances however often take their toll on our ecosystem, with high demands on energy, contributions to greenhouse gas emissions and severe environmental pollution. Mitigating these adverse effects is amongst the grand challenges of our society and at the forefront of materials research.
The currently emerging forms of soft, biologically inspired electronics and robotics have the unique potential of becoming not only like their natural antitypes in performance and capabilities, but also in terms of their ecological footprint. This talk introduces materials and methods or soft systems that facilitate a broad range of applications, from sustainably powered, transient electronic skins to metabolizable soft robots. Based on highly stretchable biogels and degradable elastomers, our forms of soft electronics and robots are designed for prolonged operation in ambient conditions without fatigue, but fully degrade after use through biological triggers. Electronic skins provide sensory feedback such as pressure maps, strain, temperature and humidity sensing. Recent advances in 3D printing of biodegradable hydrogels enables omnidirectional soft robots with multifaceted optical sensing abilities. wearable sweat sensors.