Designs of champion-level systems dedicated to a game have been considered as milestones for Artificial Intelligence. Such a success has not yet happened for the game of Bridge because (i) Bridge is a partially observ...
详细信息
ISBN:
(数字)9783319999609
ISBN:
(纸本)9783319999609;9783319999593
Designs of champion-level systems dedicated to a game have been considered as milestones for Artificial Intelligence. Such a success has not yet happened for the game of Bridge because (i) Bridge is a partially observable game (ii) a Bridge player must be able to explain at some point the meaning of his actions to his opponents. this paper presents a simple supervised learning problem in Bridge: given a 'limit hand', should a player bid or not, only considering his hand and the context of his decision. We describe this problem and some of its candidate modelisations. We then experiment state of the art propositional machine learning and ILP systems on this problem. Results of these preliminary experiments show that ILP systems are competitive or even outperform propositional Machine Learning systems. ILP systems are moreover able to build explicit models that have been validated by expert Bridge players.
Boolean networks (and more general logic models) are useful frameworks to study signal transduction across multiple pathways. logic models can be learned from a prior knowledge network structure and multiplex phosphop...
详细信息
Boolean networks (and more general logic models) are useful frameworks to study signal transduction across multiple pathways. logic models can be learned from a prior knowledge network structure and multiplex phosphoproteomics data. However, most efficient and scalable training methods focus on the comparison of two time-points and assume that the system has reached an early steady state. In this paper, we generalize such a learning procedure to take into account the time series traces of phosphoproteomics data in order to discriminate Boolean networks according to their transient dynamics. To that end, we identify a necessary condition that must be satisfied by the dynamics of a Boolean network to be consistent with a discretized time series trace. Based on this condition, we use Answer Set programming to compute an over-approximation of the set of Boolean networks which fit best with experimental data and provide the corresponding encodings. Combined with model-checking approaches, we end up with a global learning algorithm. Our approach is able to learn logic models with a true positive rate higher than 78% in two case studies of mammalian signaling networks;for a larger case study, our method provides optimal answers after 7 min of computation. We quantified the gain in our method predictions precision compared to learning approaches based on static data. Finally, as an application, our method proposes erroneous time-points in the time series data with respect to the optimal learned logic models. (C) 2016 Elsevier Ireland Ltd. All rights reserved.
A tableau calculus for a logic with constructive negation and an implementation of the related decision procedure is presented. this logic is an extension of Nelson logic and it has been used in the framework of progr...
详细信息
ISBN:
(纸本)3540230246
A tableau calculus for a logic with constructive negation and an implementation of the related decision procedure is presented. this logic is an extension of Nelson logic and it has been used in the framework of program verification and timing analysis of combinatorial circuits. the decision procedure is tailored to shrink the search space of proofs and it is proved correct by using a semantical technique. It has been implemented in C++ language.
the proceedings contain 29 papers. the special focus in this conference is on Integration of AI and OR Techniques in Constraint programming. the topics include: On CNF encodings of decision diagrams;time-series constr...
ISBN:
(纸本)9783319339535
the proceedings contain 29 papers. the special focus in this conference is on Integration of AI and OR Techniques in Constraint programming. the topics include: On CNF encodings of decision diagrams;time-series constraints: improvements and application in CP and MIP contexts;decomposition based on decision diagrams;logic-based decomposition methods for the travelling purchaser problem;lagrangian decomposition via sub-problem search;non-linear optimization of business models in the electricity market;weighted spanning tree constraint with explanations;application to an energy cost-aware production planning problem for tissue manufacturing;cyclic routing of unmanned aerial vehicles;parallelizing constraint programming with learning;parallel composition of scheduling solvers;rail capacity modelling with constraint programming;scheduling home hospice care withlogic-based benders decomposition;a global constraint for mining sequential patterns with GAP constraint;a reservoir balancing constraint with applications to bike-sharing;optimization models for a real-world snow plow routing problem;a stochastic continuous optimization backend for MiniZinc with applications to geometrical placement problems;balancing nursing workload by constraint programming;designing spacecraft command loops using two-dimension vehicle routing;constraint programming approach for spatial packaging problem;detecting semantic groups in MIP Models;revisiting two-sided stability constraints;optimal flood mitigation over flood propagation approximations;a bit-vector solver with word-level propagation;a new solver for the minimum weighted vertex cover problem and optimal upgrading schemes for effective shortest paths in networks.
We consider the problem of learning Boltzmann machine classifiers from relational data. Our goal is to extend the deep belief framework of RBMs to statistical relational models. this allows one to exploit the feature ...
详细信息
ISBN:
(数字)9783319780900
ISBN:
(纸本)9783319780900;9783319780894
We consider the problem of learning Boltzmann machine classifiers from relational data. Our goal is to extend the deep belief framework of RBMs to statistical relational models. this allows one to exploit the feature hierarchies and the non-linearity inherent in RBMs over the rich representations used in statistical relational learning (SRL). Specifically, we use lifted random walks to generate features for predicates that are then used to construct the observed features in the RBM in a manner similar to Markov logic Networks. We show empirically that this method of constructing an RBM is comparable or better than the state-of-the-art probabilistic relational learning algorithms on six relational domains.
Programmable logic Controllers (PLCs) play a significant role in the control of production systems and Sequential Function Chart (SFC) is one of the main programming languages. the reaction time of a PLC is a fundamen...
详细信息
ISBN:
(纸本)9781424415052
Programmable logic Controllers (PLCs) play a significant role in the control of production systems and Sequential Function Chart (SFC) is one of the main programming languages. the reaction time of a PLC is a fundamental matter in discrete event control systems. We will show that the reaction time of PLC depends greatly on the SFC structure, on the events sequence and also on the algorithm that executes the SFC. Five algorithms have been analyzed: Brute Force, Enabled Transitions, Representing Places, Deferred Transit and the Immediate Transit evolution models. the analysis has been carried out over a SFC library composed by well known models which can be scaled using a parameter. Finally we propose a new SFC execution technique adapted to efficiently execute a subclass of SFCs. We call this technique the Active Steps Algorithm.
In a previous work, the first author extended to higher-order rewriting and dependent types the use of size annotations in types, a termination proof technique called type or size based termination and initially devel...
详细信息
ISBN:
(纸本)3540482814
In a previous work, the first author extended to higher-order rewriting and dependent types the use of size annotations in types, a termination proof technique called type or size based termination and initially developed for ML-like programs. Here, we go one step further by considering conditional rewriting and explicit quantifications and constraints on size annotations. this allows to describe more precisely how the size of the output of a function depends on the size of its inputs. Hence, we can check the termination of more functions. We first give a general type-checking algorithm based on constraint solving. then, we give a termination criterion with constraints in Presburger arithmetic. To our knowledge, this is the first termination criterion for higher-order conditional rewriting taking into account the conditions in termination.
To investigate the impact of the occurrence of a phase transition (PT) in the covering test on the learning success rate, systematic experiments with several learning algorithms have been conducted on a large set of a...
ISBN:
(纸本)9783540738466
To investigate the impact of the occurrence of a phase transition (PT) in the covering test on the learning success rate, systematic experiments with several learning algorithms have been conducted on a large set of artificially generated problems by Botta et al. [3]. the authors generated a set of 451 problems by choosing each target concept according to its location in the (m,L) plane with respect to the PT. the “yes”, “no” and “pt” regions are uniformly visited by varying (m,L) pairs without replacement (m ranges in [5,30] and L ranges in [12,40]). One important conclusion of their work is that the occurrence of a PT in the covering test is a general problem for all learning algorithms: the PT is viewed as an attractor for the heuristic search of any learning algorithms, which are bound to find a concept definition in the PT. Moreover, for all tested learners, there exists a failure region, starting from the “pt” region to the beginning of the “no” region, where the learnt theories are seemingly randomly constructed, with no better predictive accuracy than random guessing.
the proceedings contain 16 papers. the special focus in this conference is on Frontiers of Combining Systems. the topics include: Vampire with a Brain Is a Good ITP Hammer;optimization Modulo Non-linear Arithmetic via...
ISBN:
(纸本)9783030862046
the proceedings contain 16 papers. the special focus in this conference is on Frontiers of Combining Systems. the topics include: Vampire with a Brain Is a Good ITP Hammer;optimization Modulo Non-linear Arithmetic via Incremental Linearization;Quantifier Simplification by Unification in SMT;algorithmic Problems in the Symbolic Approach to the Verification of Automatically Synthesized Cryptosystems;formal Analysis of Symbolic Authenticity;Formal Verification of a Java Component Using the RESOLVE Framework;non-disjoint Combined Unification and Closure by Equational Paramodulation;symbol Elimination and Applications to Parametric Entailment Problems;on the Copy Complexity of Width 3 Horn Constraint Systems;Restricted Unification in the DL FL0;combining Event Calculus and Description logic Reasoning via logicprogramming;semantic Forgetting in Expressive Description logics;improving Automation for Higher-Order Proof Steps;JEFL: Joint Embedding of Formal Proof Libraries.
It is well-known that heuristic search in ILP is prone to plateau phenomena. An explanation can be given after the work of Giordana and Saitta: the ILP covering test is NP-complete and therefore exhibits a sharp phase...
详细信息
It is well-known that heuristic search in ILP is prone to plateau phenomena. An explanation can be given after the work of Giordana and Saitta: the ILP covering test is NP-complete and therefore exhibits a sharp phase transition in its coverage probability. As the heuristic value of a hypothesis depends on the number of covered examples, the regions "yes" and "no" represent plateaus that need to be crossed during search without an informative heuristic value. Several subsequent works have extensively studied this finding by running several learning algorithms on a large set of artificially generated problems and argued that the occurrence of this phase transition dooms every learning algorithm to fail to identify the target concept. We note however that only generate-and-test learning algorithms have been applied and that this conclusion has to be qualified in the case of data-driven learning algorithms. Mostly building on the pioneering work of Winston on near-miss examples, we show that, on the same set of problems, a top-down data-driven strategy can cross any plateau if near-misses are supplied in the training set, whereas they do not change the plateau profile and do not guide a generate-and-test strategy. We conclude that the location of the target concept with respect to the phase transition alone is not a reliable indication of the learning problem difficulty as previously thought.
暂无评论