We have relevant datasets, repositories, frameworks and tools of relevance for research and technology transfer initiatives related to knowledge extraction. This section provides an overview on a selection of them and links to download or contact details.

The MdM Strategic Research Program has its own community in Zenodo for material available in this repository  as well as at the UPF e-repository  . Below a non-exhaustive list of datasets representative of the research in the Department.

As part of the promotion of the availability of resources, the creation of specific communities in Zenodo has also been promoted, at level of research communities (for instance, MIR and Educational Data Analytics) or MSc programs (for instance, the Master in Sound and Music Computing)

 

 

Back [MSc thesis] Cross-Entropy method for Kullback-Leibler control in multi-agent systems

[MSc thesis] Cross-Entropy method for Kullback-Leibler control in multi-agent systems

Author: Beatriz Cabrero Daniel

Supervisor: Mario Ceresa, Vicenç Gómez

MSc program: Master in Intelligent Interactive Systems

We consider the problem of computing optimal control policies in large-scale multiagent systems, for which the standard approach via the Bellman equation is intractable. Our formulation is based on the Kullback-Leibler control framework, also known as Linearly-Solvable Markov Decision Problems. In this setting, adaptive importance sampling methods have been derived that, when combined with function approximation, can be effective for high-dimensional systems. Our approach iteratively learns an importance sampler from which the optimal control can be extracted and requires to simulate and reweight agents’ trajectories in the world multiple times. We illustrate our approach through a modified version of the popular stag-hunt game; in this scenario, there is a multiplicity of optimal policies depending on the “temperature” parameter of the environment. The system is built inside Pandora, a multi-agent-based modeling framework and toolbox for parallelization, freeing us from dealing with memory management when running multiple simulations. By using function approximation and assuming some particular factorization of the system dynamics, we are able to scale-up our method to problems with M = 12 agents moving in two-dimensional grids of size N = 21×21, improving on existing methods that perform approximate inference on a temporal probabilistic graphical model.

Additional material:

 

Link: http://hdl.handle.net/10230/33109

Related Assets