BoolVar/PB is an open source java library dedicated to the translation of pseudo-Boolean constraints into CNF formulae. Input constraints can be categorized with tags. Several encoding schemes are implemented in a way that each input constraint can be translated using one or several encoders, according to the related tags. The library can be easily extended by adding new encoders and / or new output formats.
Building boolean satisfiability solvers that implement strong proof systems is an important goal for the field of AI. This goal is motivated by results from the field of proof complexity showing that some proof systems are more limited than others. An inference system is limited if it is impossible to construct short proofs of unsatisfiability for certain families of problems. These results have significant consequences for systematic satisfiability solvers. Because systematic solvers can be viewed primarily as constructing proofs of unsatisfiability, it follows that these solvers are subject to the limitations of the proof systems they implement.
The class of Gaussian Process (GP) methods for Temporal Difference learning has shown promise for data-efficient model-free Reinforcement Learning. In this paper, we consider a recent variant of the GP-SARSA algorithm, called Sparse Pseudo-input Gaussian Process SARSA (SPGP-SARSA), and derive recursive formulas for its predictive moments. This extension promotes greater memory efficiency, since previous computations can be reused and, interestingly, it provides a technique for updating value estimates on a multiple timescales
Bayesian inference in the presence of an intractable likelihood function is computationally challenging. When following a Markov chain Monte Carlo (MCMC) approach to approximate the posterior distribution in this context, one typically either uses MCMC schemes which target the joint posterior of the parameters and some auxiliary latent variables or pseudo-marginal Metropolis-Hastings (MH) schemes which mimic a MH algorithm targeting the marginal posterior of the parameters by approximating unbiasedly the intractable likelihood. In scenarios where the parameters and auxiliary variables are strongly correlated under the posterior and/or this posterior is multimodal, Gibbs sampling or Hamiltonian Monte Carlo (HMC) will perform poorly and the pseudo-marginal MH algorithm, as any other MH scheme, will be inefficient for high dimensional parameters. We propose here an original MCMC algorithm, termed pseudo-marginal HMC, which approximates the HMC algorithm targeting the marginal posterior of the parameters. We demonstrate through experiments that pseudo-marginal HMC can outperform significantly both standard HMC and pseudo-marginal MH schemes.
Unit propagation-based (UP) lower bounds are used in the vast majority of current Max-SAT solvers. However, lower bounds based on UP have seldom been applied in Pseudo-Boolean Optimization (PBO) algorithms derived from the DPLL procedure for Propositional Satisfiability (SAT). This paper enhances a DPLL-style PBO algorithm with an UP lower bound, and establishes conditions that enable constraint learning and non-chronological backtracking in the presence of conflicts involving constraints generated by the UP lower bound. From a theorical point of view, the paper highlights the relationship between the recent UP lower bound and the well-known Maximum Independent Set (MIS) lower bound. Finally, the paper provides preliminary results that show the effectiveness of the proposed approach for representative sets of instances.