In this work we use deep reinforcement learning to create an autonomous agent that can navigate in a two-dimensional space using only raw auditory sensory information from the environment, a problem that has received very little attention in the reinforcement learning literature. Our experiments show that the agent can successfully identify a particular target speaker among a set of N predefined speakers in a room and move itself towards that speaker, while avoiding collision with other speakers or going outside the room boundaries. The agent is shown to be robust to speaker pitch shifting and it can learn to navigate the environment, even when a limited number of training utterances are available for each speaker.

No abstract

No abstract

Abstract. The primary objective of SIRENE is to simulate the response to neutrino events of any type of high energy neutrino telescope. Additionally, it implements different geometries for a neutrino detector and different configurations and characteristics of photomultiplier tubes (PMTs) inside the optical modules of the detector through a library of C++ classes. This could be considered a massive statistical analysis of photo-electrons. Aim of this work is the development of a multithreaded version of the SIRENE detector simulation software for high energy neutrinos. This approach allows utilization of multiple CPU cores leading to a potentially significant decrease in the required execution time compared to the sequential code. We are making use of the OpenMP framework for the production of multithreaded code running on the CPU. Finally, we analyze the feasibility of a GPUaccelerated implementation.

No abstract