Balázs Pintér

Towards a Smart Wearable Tool to Enable People with SSPI to Communicate by Sentence Fragments

The ability to communicate with others is of paramount importance for mental well-being. In this paper, we describe an interaction system to reduce communication barriers for people with severe speech and physical impairments (SSPI) such as cerebral palsy. The system consists of two main components: (i) the head-mounted human-computer interaction (HCI) part consisting of smart glasses with gaze trackers and text-to-speech functionality (which implement a communication board and the selection tool), and (ii) a natural language processing pipeline in the backend in order to generate complete sentences from the symbols on the board. We developed the components to provide a smooth interaction between the user and the system thereby including gaze tracking, symbol selection, symbol recognition, and sentence generation. Our results suggest that such systems can dramatically increase communication efficiency of people with SSPI.

Multiagent reinforcement learning model for the emergence of common property and transhumance in sub-saharan africa

We consider social phenomena as challenges and measures for learning in multi-agent scenarios for the following reasons: (i) social phenomena emerge through complex learning processes of groups of people, (ii) a model of a phenomenon sheds light onto the strengths and weaknesses of the learning algorithm in the context of the model environment. In this paper we use tabular reinforcement learning to model the emergence of common property and transhumance in Sub-Saharan Africa. We find that the Markovian assumption is sufficient for the emergence of property sharing, when (a) the availability of resources fluctuates (b) the agents try to maximize their resource intake independently and (c) all agents learn simultaneously.