the proceedings contain 14 papers. the special focus in this conference is on inductivelogicprogramming. the topics include: Relational kernel-based grasping with numerical features;complex aggregates within random ...
ISBN:
(纸本)9783319405650
the proceedings contain 14 papers. the special focus in this conference is on inductivelogicprogramming. the topics include: Relational kernel-based grasping with numerical features;complex aggregates within random forests;distributed parameter learning for probabilistic ontologies;meta-interpretive learning of data transformation programs;statistical relational learning with soft quantifiers;ontology learning from interpretations in lightweight description logics;constructing markov logic networks from first-order default rules;a note on mining all graphs;processing markov logic networks with GPUs;using ilp to identify pathway activation patterns in systems biology;an algebraic prolog for kernel programming;an exercise in declarative modeling for relational query mining;learning inference by induction and identification of transition models of biological systems in the presence of transition noise.
Weather forecasting is important for saving lives, protecting property, and supporting economic activities. It provides timely warnings for severe weather, improves agricultural planning, and aids in disaster manageme...
详细信息
ISBN:
(纸本)9783031742088;9783031742095
Weather forecasting is important for saving lives, protecting property, and supporting economic activities. It provides timely warnings for severe weather, improves agricultural planning, and aids in disaster management. Neural networks and deep learning methods can achieve impressive accuracy in weather prediction, but their black-box nature lacks in explainability. To address this limitation, we investigated the potential of FastLAS, an inductivelogicprogramming (ilp) framework, to produce reliable and, more important, explainable weather predictions. FastLAS learns ASP programs whose syntax and structural semantics resemble natural human language, making them easily understandable and interpretable by humans. the supportedness of stable models allows a clear explanation of the predictions. Our empirical evaluation on data from an Italian weather forecasting center shows that our approach is capable of learning predictive models from small dataset (a few samples instead of the thousands needed by neural networks) achieving an accuracy higher than statistical machine learning base lines.
the proceedings contain 25 papers. the special focus in this conference is on inductivelogicprogramming. the topics include: A personal view of how best to apply ilp;agents that reason and learn;mining model trees;c...
ISBN:
(纸本)9783540399179
the proceedings contain 25 papers. the special focus in this conference is on inductivelogicprogramming. the topics include: A personal view of how best to apply ilp;agents that reason and learn;mining model trees;complexity parameters for first-order classes;a multi-relational decision tree learning algorithm;applying theory revision to the design of distributed databases;disjunctive learning with a soft-clustering method;ilp for mathematical discovery;an exhaustive matching procedure for the improvement of learning efficiency;efficient data structures for inductivelogicprogramming;graph kernels and gaussian processes for relational reinforcement learning;on condensation of a clause;a comparative evaluation of feature set evolution strategies for multi-relational boosting;comparative evaluation of approaches to propositionalization;improved distances for structured data;induction of enzyme classes from biological databases;estimating maximum likelihood parameters for stochastic context-free graph grammars;induction of the effects of actions by monotonic methods;hybrid abductive inductive learning;query optimization in inductivelogicprogramming by reordering literals;efficient learning of unlabeled term trees with contractible variables from positive data;relational IBL in music with a new structural similarity measure and an effective grammar-based compression algorithm for tree structured data.
My research explores integrating deep learning and logicprogramming to set the basis for a new generation of AI systems. By combining neural networks withinductivelogicprogramming (ilp), the goal is to construct s...
详细信息
My research explores integrating deep learning and logicprogramming to set the basis for a new generation of AI systems. By combining neural networks withinductivelogicprogramming (ilp), the goal is to construct systems that make accurate predictions and generate comprehensible rules to validate these predictions. Deep learning models process and analyze complex data, while ilp techniques derive logical rules to prove the network's conclusions. Explainable AI methods, like eXplainable Answer Set programming (XASP), elucidate the reasoning behind these rules and decisions. the focus is on applying ilp frameworks, specifically ILASP and FastLAS, to enhance explainability in various domains. My test cases span weather prediction, the legal field, and image recognition. In weather forecasting, the system will predict events and provides explanations using FastLAS, with plans to integrate recurrent neural networks in the future. In the legal domain, the research focuses on interpreting vague decisions and assisting legal professionals by encoding Italian legal articles and learning reasoning patterns from Court of Cassation decisions using ILASP. For biological laboratories, we will collaborate with a research group to automate spermatozoa morphology classification for Bull Breeding Soundness Evaluation using YOLO networks and ilp to explain classification outcomes. this hybrid approach aims to bridge the gap between the high performance of deep learning models and the transparency of symbolic reasoning, advancing AI by providing interpretable and trustworthy applications.
the proceedings contain 14 papers. the special focus in this conference is on inductivelogicprogramming. the topics include: Reframing on relational data;inductive learning using constraint-driven bias;nonmonotonic ...
ISBN:
(纸本)9783319237077
the proceedings contain 14 papers. the special focus in this conference is on inductivelogicprogramming. the topics include: Reframing on relational data;inductive learning using constraint-driven bias;nonmonotonic learning in large biological networks;construction of complex aggregates with random restart hill-climbing;logical minimisation of meta-rules within meta-interpretive learning;goal and plan recognition via parse trees using prefix and infix probability computation;effectively creating weakly labeled training examples via approximate domain knowledge;learning prime implicant conditions from interpretation transition;statistical relational learning for handwriting recognition;the most probable explanation for probabilistic logic programs with annotated disjunctions;towards machine learning of predictive models from ecological data;pagerank, proPPR, and stochastic logic programs;complex aggregates over clusters of elements and on the complexity of frequent subtree mining in very simple structures.
Motivated by an analogy with matrix factorization, we introduce the problem of factorizing relational data. In matrix factorization, one is given a matrix and has to factorize it as a product of other matrices. In rel...
详细信息
Motivated by an analogy with matrix factorization, we introduce the problem of factorizing relational data. In matrix factorization, one is given a matrix and has to factorize it as a product of other matrices. In relational data factorization, the task is to factorize a given relation as a conjunctive query over other relations, i.e., as a combination of natural join operations. Given a conjunctive query and the input relation, the problem is to compute the extensions of the relations used in the query. thus, relational data factorization is a relational analog of matrix factorization;it is also a form of inverse querying as one has to compute the relations in the query from the result of the query. the result of relational data factorization is neither necessarily unique nor required to be a lossless decomposition of the original relation. therefore, constraints can be imposed on the desired factorization and a scoring function is used to determine its quality (often similarity to the original data). Relational data factorization is thus a constraint satisfaction and optimization problem. We show how answer set programming can be used for solving relational data factorization problems.
A distributed system DS can be formalized as a state machine M and many desired properties of DS can be expressed as invariants of M. An invariant of M is a state predicate p of M such that p holds for all reachable s...
详细信息
We study planning in relational Markov decision processes involving discrete and continuous states and actions, and an unknown number of objects. this combination of hybrid relational domains has so far not received a...
详细信息
We study planning in relational Markov decision processes involving discrete and continuous states and actions, and an unknown number of objects. this combination of hybrid relational domains has so far not received a lot of attention. While both relational and hybrid approaches have been studied separately, planning in such domains is still challenging and often requires restrictive assumptions and approximations. We propose HYPE: a sample-based planner for hybrid relational domains that combines model-based approaches with state abstraction. HYPE samples episodes and uses the previous episodes as well as the model to approximate the Q-function. In addition, abstraction is performed for each sampled episode, this removes the complexity of symbolic approaches for hybrid relational domains. In our empirical evaluations, we show that HYPE is a general and widely applicable planner in domains ranging from strictly discrete to strictly continuous to hybrid ones, handles intricacies such as unknown objects and relational models. Moreover, empirical results showed that abstraction provides significant improvements.
Expert knowledge can often be represented using default rules of the form "if A then typically B". In a probabilistic framework, such default rules can be seen as constraints on what should be derivable by M...
详细信息
ISBN:
(数字)9783319405667
ISBN:
(纸本)9783319405667;9783319405650
Expert knowledge can often be represented using default rules of the form "if A then typically B". In a probabilistic framework, such default rules can be seen as constraints on what should be derivable by MAP-inference. We exploit this idea for constructing a Markov logic network M from a set of first-order default rules D, such that MAP inference from M exactly corresponds to default reasoning from D, where we view first-order default rules as templates for the construction of propositional default rules. In particular, to construct appropriate Markov logic networks, we lift three standard methods for default reasoning. the resulting Markov logic networks could then be refined based on available training data. Our method thus offers a convenient way of using expert knowledge for constraining or guiding the process of learning Markov logic networks.
暂无评论