1 code implementation • 17 Dec 2021 • Florent Delgrange, Ann Nowé, Guillermo A. Pérez
Finally, we show how one can use a policy obtained via state-of-the-art RL to efficiently train a variational autoencoder that yields a discrete latent model with provably approximately correct bisimulation guarantees.
1 code implementation • 22 Mar 2023 • Florent Delgrange, Ann Nowé, Guillermo A. Pérez
Our approach yields bisimulation guarantees while learning the distilled policy, allowing concrete optimization of the abstraction and representation model quality.
1 code implementation • 26 Nov 2016 • Krishnendu Chatterjee, Petr Novotný, Guillermo A. Pérez, Jean-François Raskin, Đorđe Žikelić
In this work we go beyond both the "expectation" and "threshold" approaches and consider a "guaranteed payoff optimization (GPO)" problem for POMDPs, where we are given a threshold $t$ and the objective is to find a policy $\sigma$ such that a) each possible outcome of $\sigma$ yields a discounted-sum payoff of at least $t$, and b) the expected discounted-sum payoff of $\sigma$ is optimal (or near-optimal) among all policies satisfying a).
no code implementations • 24 Apr 2018 • Jan Křetínský, Guillermo A. Pérez, Jean-François Raskin
Assuming the support of the unknown transition function and a lower bound on the minimal transition probability are known in advance, we show that in MDPs consisting of a single end component, two combinations of guarantees on the parity and mean-payoff objectives can be achieved depending on how much memory one is willing to use.
no code implementations • 13 Jul 2018 • Nikhil Balaji, Stefan Kiefer, Petr Novotný, Guillermo A. Pérez, Mahsa Shirmohammadi
We show that, given a horizon $n$ in binary and an MDP, computing an optimal policy is EXP-complete, thus resolving an open problem that goes back to the seminal 1987 paper on the complexity of MDPs by Papadimitriou and Tsitsiklis.
no code implementations • 17 Nov 2018 • Michaël Cadilhac, Guillermo A. Pérez, Marie van den Bogaard
Discounted-sum games provide a formal model for the study of reinforcement learning, where the agent is enticed to get rewards early since later rewards are discounted.
no code implementations • 6 Apr 2020 • Floris Geerts, Filip Mazowiecki, Guillermo A. Pérez
In this paper we cast neural networks defined on graphs as message-passing neural networks (MPNNs) in order to study the distinguishing power of different classes of such models.
no code implementations • 12 May 2020 • Dennis Gross, Nils Jansen, Guillermo A. Pérez, Stephan Raaijmakers
The robustness-checking problem consists of assessing, given a set of classifiers and a labelled data set, whether there exists a randomized attack that induces a certain expected loss against all classifiers.
no code implementations • 19 May 2020 • Damien Busatto-Gaston, Debraj Chakraborty, Shibashis Guha, Guillermo A. Pérez, Jean-François Raskin
In this paper, we investigate the combination of synthesis, model-based learning, and online sampling techniques to obtain safe and near-optimal schedulers for a preemptible task scheduling problem.
no code implementations • 28 Jan 2021 • Michael Blondin, Tim Leys, Filip Mazowiecki, Philip Oftermatt, Guillermo A. Pérez
Our three main results are as follows: (1) We prove that the reachability problem for COCA with global upper and lower bound tests is in NC2; (2) that, in general, the problem is decidable in polynomial time; and (3) that it is decidable in the polynomial hierarchy for COCA with parametric counter updates and bound tests.
Formal Languages and Automata Theory Logic in Computer Science
no code implementations • 3 May 2020 • Guillermo A. Pérez, Ritam Raha
We study the (parameter) synthesis problem for one-counter automata with parameters.
Logic in Computer Science Formal Languages and Automata Theory
no code implementations • 6 Mar 2023 • Raphael Avalos, Florent Delgrange, Ann Nowé, Guillermo A. Pérez, Diederik M. Roijers
Maintaining a probability distribution that models the belief over what the true state is can be used as a sufficient statistic of the history, but its computation requires access to the model of the environment and is often intractable.
no code implementations • 9 May 2023 • Kasper Engelen, Guillermo A. Pérez, Shrisha Rao
In terms of computational complexity, we establish that determining whether p is never worse than q is coETR-complete.
no code implementations • 15 Aug 2023 • Debraj Chakraborty, Damien Busatto-Gaston, Jean-François Raskin, Guillermo A. Pérez
In particular, we use model-checking techniques to guide the MCTS algorithm in order to generate offline samples of high-quality decisions on a representative set of states of the MDP.
no code implementations • 21 Feb 2024 • Florent Delgrange, Guy Avni, Anna Lukina, Christian Schilling, Ann Nowé, Guillermo A. Pérez
We propose a novel approach to the problem of controller design for environments modeled as Markov decision processes (MDPs).
no code implementations • 4 Mar 2024 • Véronique Bruyère, Bharat Garhewal, Guillermo A. Pérez, Gaëtan Staquet, Frits W. Vaandrager
However, whereas Waga needs exponentially many concrete queries to implement a single symbolic query, we only need a polynomial number.