AIIA 2019

Abstract. Dynamic systems that operate autonomously in nondeterministic (uncertain) environments are becoming a reality. These include intelligent robots, self-driving cars, but also manufacturing systems (Industry 4.0), smart objects and spaces (IoT), advanced business process management systems (BPM), and many others. These systems are currently being revolutionized by advancements in sensing (vision, language understanding) and actuation components (autonomous mobile manipulators, automated storage and retrieval systems). However, in spite of these advances, their core logic is still mainly based on hard-wired rules either designed or possibly obtained through a learning process.

On the other hand, we can envision systems that are able to deliberate by themselves about their course of action when un-anticipated circumstances arise, new goals are submitted, new safety conditions are required, and new regulations and conventions are imposed. Crucially, empowering dynamic systems with deliberating capabilities carries significant risks and therefore we must be able to balance such power with trust. For this reason it is of interest to make these systems queryable, analyzable and explainable in human terms, so as to be guarded by human oversight.

In this talk we discuss how recent scientific discoveries in Knowledge Representation and Planning combined with insights from Verification and Synthesis in Formal Methods, Data-Aware Processes in Databases, as well as other areas of AI, chart a novel path for realizing what we may call Queryable Self-Deliberating Dynamic Systems. That is, systems with a multifaceted model of the world that can be exploited to deliberate on their course of action and answer queries about their behavior.

Abstract. AI is currently dominated by data science, a six decades old field that began with the first attempt to build a learning machine by Arthur Samuel to play checkers on an IBM 701 computer in 1959. AI has become commercially highly successful, with economic impact estimated to be in the trillions of dollars. As AI scales, however, to address global problems, the weaknesses of relying extensively on data science are becoming more apparent. The computational costs of deep learning have grown by a factor of 500,000 in the past 5 years. The resulting models are opaque and hard to interpret. The requirements of collecting and labeling huge datasets are difficult to satisfy in many areas.

In this talk, I will outline a new vision for AI, one that is based on imagination. Data science concerns itself with statistical summarization of the past history. Imagination science is concerned with understanding the future, of evaluating the impact of interventions, of causal and counterfactual reasoning, and analogical reasoning and creativity. Imagination has been viewed as a uniquely human capability, one that allows us to compose symphonies, create abstract art, imagine alien life in distant galaxies, invent new technologies and model the large-scale structure of the universe. I will sketch out a few lines of research that are providing the first tangible steps on building imagination machines.

Abstract. Designing good models is one of the main challenges for obtaining realistic and useful decision support and optimization systems. Traditionally combinatorial models are crafted by interacting with domain experts with limited accuracy guarantees. Nowadays we have access to data sets of unprecedented scale and accuracy about the systems we are deciding on.

In this talk we propose a methodology called Empirical Model Learning that uses machine learning to extract data-driven decision model components and integrates them into an expert-designed decision model. We outline the main domains where EML could be useful and we show how to ground Empirical Model Learning on a problem of thermal-aware workload allocation and scheduling on a multi-core platform.

In addition, we discuss how to use EML with different optimization and machine learning techniques, and we provide some hints about recent work on EML for hierarchical optimization and on-line/off-line optimization.

Abstract. The world around us is highly complex but Autonomous Systems must be able to reliably make accurate decisions that in many cases may even affect human lives. With Digital Reality we propose an approach that instead of only relying on real data, learns models of the real world and uses synthetic sensor data generated via simulations, for the training and -- even more importantly -- the validation of Autonomous Systems. This is extended by a continuous process of validating the models against the real world for improving and adapting them to a changing environment.

A highly relevant application of this approach is in intelligent sensor systems. Using a model about the object to be measured and the measuring process these systems are aware of what and how they are measuring and can adapt the measuring strategy and parameters accordingly, e.g. to obtain accurate measurements or target high throughput.