检索结果-内蒙古大学图书馆

Consensus-based modeling using distributed feature construction with ILP

MACHINE LEARNING 2018年第5期107卷 825-858页

作者： Dutta, Haimonti Srinivasan, Ashwin SUNY Buffalo Dept Management Sci & Syst Buffalo NY 14260 USA BITS Pilani Dept Comp Sci & Informat Syst Goa Campus Pilani Goa India

A particularly successful role for inductive logic programming (ILP) is as a tool for discovering useful relational features for subsequent use in a predictive model. Conceptually, the case for using ILP to construct relational features rests on treating these features as functions, the automated discovery of which necessarily requires some form of first-order learning. Practically, there are now several reports in the literature that suggest that augmenting any existing feature with ILP-discovered relational features can substantially improve the predictive power of a model. While the approach is straightforward enough, much still needs to be done to scale it up to explore more fully the space of possible features that can be constructed by an ILP system. This is in principle, infinite and in practice, extremely large. Applications have been confined to heuristic or random selections from this space. In this paper, we address this computational difficulty by allowing features and models to be constructed in a distributed manner. That is, there is a network of computational units, each of which employs an ILP engine to construct some small number of features and then builds a (local) model. We then employ an asynchronous consensus-based algorithm, in which neighboring nodes share information and update local models. This gossip-based information exchange results in the formation of non-stationary Markov chains. For a category of models (those with convex loss functions), it can be shown (using the Supermartingale Convergence Theorem) that the algorithm will result in all nodes converging to a consensus model. In practice, it may be slow to achieve this convergence. Nevertheless, our results on synthetic and real datasets suggest that in relatively short time the "best" node in the network reaches a model whose predictive accuracy is comparable to that obtained using more computational effort in a non-distributed setting (the best node is identified as the one whose

关键词： inductive logic programming Consensus based learning Stochastic gradient descent Feature selection

来源：评论

学校读者我要写书评

暂无评论

ILP-assisted de novo drug design

引用

MACHINE LEARNING 2016年第3期103卷 309-341页

作者： Kaalia, Rama Srinivasan, Ashwin Kumar, Amit Ghosh, Indira Jawaharlal Nehru Univ Sch Computat & Integrat Sci New Delhi 110067 India BITS Dept Comp Sci & Informat Syst Pilani Goa India

De novo design of drugs uses the three-dimensional structure of a target protein (often called the receptor) to design molecules (or ligands) that could bind to the receptor and hence inhibit its functioning. Thus, unlike a ligand-based approach, this form of drug design does not require prior knowledge of inhibitors. In this paper, the three-dimensional structure of a receptor is used indirectly, in the form of molecular interaction fields of the receptor and small molecules (or probes). In addition, we also use domain-specific constraints encoding basic geometric and pharmacological requirements imposed by the target. Interaction energies of one or more targets with a set of probes are used to identify three-dimensional constraints that occur in many-preferably all-targets. In a graph-theoretic sense, the constraints are (small, fixed-size) cliques in graphs with labelled vertices representing probe-specific points of high interaction energy, and edges between a pair of vertices are labelled by the three-dimensional distance between the corresponding points of interaction. Our interest is in the discovery of frequent cliques that satisfy domain-specific constraints. In the paper, the discovery of such patterns is done using an inductive logic programming (ILP) engine. The case for the use of ILP stems primarily from the explicit ways of incorporating domain-constraints, but any other technique capable of discovering frequent cliques from data can be used with some additional effort. The frequent cliques discovered are used to hypothesize pharmacophore-like structures on potential ligands. We test the utility of this approach by conducting a case study on the discovery of anti-malarials. Specifically, we test the approach on proteins belonging to the class of aspartic proteases. We are particularly interested in plasmepsin II, which is an enzyme in the haemoglobin degradation pathway of Plasmodium falciparum. We assess the pharmacophore-like constraints using: (a)

关键词： De novo drug design inductive logic programming Pharmacophores Frequent cliques

来源：评论

学校读者我要写书评

暂无评论

Sequential inference with reliable observations: Learning to construct force-dynamic models

引用

ARTIFICIAL INTELLIGENCE 2006年第14-15期170卷 1081-1100页

作者： Fern, Alan Givan, Robert Oregon State Univ Sch Elect Engn & Comp Sci Corvallis OR 97331 USA Purdue Univ Sch Elect & Comp Engn W Lafayette IN 47907 USA

We present a trainable sequential-inference technique for processes with large state and observation spaces and relational structure. We apply our technique to the problem of force-dynamic state inference from video, which is a critical component of the LEONARD [J.M. Siskind, Grounding lexical semantics of verbs in visual perception using force dynamics and event logic, Journal of Artificial Intelligence Research 15 (2001) 31-90] visual-event recognition system. LEONARD uses event definitions that are grounded in force-dynamic primitives-making robust and efficient force-dynamic inference critical to good performance. Our sequential-inference method assumes "reliable observations", i.e., that each process state (e.g., force-dynamic state) persists long enough to be reliably inferred from the observations (e.g., video frames) it generates. We introduce the idea of a "state-inference function" (from observation sequences to underlying hidden states) for representing knowledge about a process and develop an efficient sequential-inference algorithm, utilizing this function, that is correct for processes that generate reliable observations consistent with the state-inference function. We describe a representation for state-inference functions in relational domains and give a corresponding supervised learning algorithm. Our experiments in force-dynamic state inference show that our technique provides significantly improved accuracy and speed relative to a variety of recent, hand-coded, non-trainable systems, and a trainable system based on probabilistic modeling. (C) 2006 Published by Elsevier B.V.

关键词： sequence learning relational learning event recognition temporal learning inductive logic programming

来源：评论

学校读者我要写书评

暂无评论

Clustering and instance based learning in first order logic

引用

AI COMMUNICATIONS 2002年第4期15卷 217-218页

作者： Ramon, J Katholieke Univ Leuven Dept Comp Sci B-3001 Heverlee Belgium

Instance based learning and clustering are popular methods in propositional machine learning. Both methods use a notion of similarity between objects. This dissertation investigates these methods in a relational setting. First, a number of new metrics are proposed. Next, these metrics are used to upgrade clustering and instance based learning to first order logic.

关键词： machine learning inductive logic programming instance based learning clustering

来源：评论

学校读者我要写书评

暂无评论

Machine learning

引用

COMPUTING & CONTROL ENGINEERING JOURNAL 1996年第4期7卷 164-168页

作者： Cussens, J Comput. Lab. Oxford Univ. UK

In this article a short review of research and applications in machine learning is given. Rather than attempt to cover all areas of ML, the focus is on its role in building expert systems, its approach to classification problems and ML methods of learning control. A relatively new area, inductive logic programming, is also discussed.

关键词： learning control learning (artificial intelligence) ML Adaptive system theory knowledge engineering Knowledge engineering techniques expert system building logic programming logic programming machine learning inductive logic programming classification problems

来源：评论

学校读者我要写书评

暂无评论

Learning Distributional Programs for Relational Autocompletion

引用

THEORY AND PRACTICE OF logic programming 2022年第1期22卷 81-114页

作者： Kumar, Nitesh Kuzelka, Ondrej De Raedt, Luc Katholieke Univ Leuven Dept Comp Sci Leuven Belgium Czech Tech Univ Dept Comp Sci Prague Czech Republic

Relational autocompletion is the problem of automatically filling out some missing values in multi-relational data. We tackle this problem within the probabilistic logic programming framework of Distributional Clauses (DCs), which supports both discrete and continuous probability distributions. Within this framework, we introduce DiceML - an approach to learn both the structure and the parameters of DC programs from relational data (with possibly missing data). To realize this, DiceML integrates statistical modeling and DCs with rule learning. The distinguishing features of DiceML are that it (1) tackles autocompletion in relational data, (2) learns DCs extended with statistical models, (3) deals with both discrete and continuous distributions, (4) can exploit background knowledge, and (5) uses an expectation-maximization-based (EM) algorithm to cope with missing data. The empirical results show the promise of the approach, even when there is missing data.

关键词： probabilistic logic programming statistical relational learning structure learning inductive logic programming

来源：评论

学校读者我要写书评

暂无评论

Statistical Relational Artificial Intelligence: From Distributions through Actions to Optimization

引用

KUNSTLICHE INTELLIGENZ 2015年第4期29卷 363-368页

作者： Kersting, Kristian Natarajan, Sriraam TU Dortmund Univ Dortmund Germany Indiana Univ Bloomington IN USA

Statistical Relational AI-the science and engineering of making intelligent machines acting in noisy worlds composed of objects and relations among the objects-is currently motivating a lot of new AI research and has tremendous theoretical and practical implications. Theoretically, combining logic and probability in a unified representation and building general-purpose reasoning tools for it has been the dream of AI, dating back to the late 1980s. Practically, successful statistical relational AI tools enable new applications in several large, complex realworld domains including those involving big data, natural text, social networks, the web, medicine and robotics, among others. Such domains are often characterized by rich relational structure and large amounts of uncertainty. logic helps to faithfully model the former while probability helps to effectively manage the latter. Our intention here is to give a brief (and necessarily incomplete) overview and invitation to the emerging field of Statistical Relational AI from the perspective of acting optimally and learning to act.

关键词： inductive logic programming logical Query Markov Decision Process Probabilistic Graphical Model Situation Calculus

来源：评论

学校读者我要写书评

暂无评论

DLP Learning from Uncertain Data

引用

Tsinghua Science and Technology 2010年第6期15卷 650-656页

作者：朱曼高志强漆桂林季秋 School of Computer Science and Engineering Southeast University

Description logic programs （DLP） are an expressive but tractable subset of OWL. This paper ana-lyzes the important under-researched problem of learning DLP from uncertain data. Current studies have rarely explored the plentiful uncertain data populating the semantic web. This algorithm handles uncertain data in an inductive logic programming framework by modifying the performance evaluation criteria. A pseudo-log-likelihood based measure is used to evaluate the performance of different literals under uncer-tainties. Experiments on two datasets demonstrate that the approach is able to automatically learn a rule-set from uncertain data with acceptable accuracy.

关键词： description logic programs inductive logic programming Markov logic networks

来源：评论

学校读者我要写书评

暂无评论

Characteristics of recovery from the euthyroid sick syndrome induced by tumor necrosis factor alpha in cancer patients

引用

METABOLISM-CLINICAL AND EXPERIMENTAL 1999年第3期48卷 324-329页

作者： Feelders, RA Swaak, AJG Romijn, JA Eggermont, AMM Tielens, ET Vreugdenhil, G Endert, E van Eijk, HG Berghout, A Univ Rotterdam Hosp Dept Internal Med 1 NL-3015 GD Rotterdam Netherlands Erasmus Univ Dept Chem Pathol Rotterdam Netherlands Zuiderziekenhuis Dept Rheumatol Rotterdam Netherlands Zuiderziekenhuis Dept Internal Med Rotterdam Netherlands Netherlands Red Cross Blood Transfus Serv Cent Lab Amsterdam Netherlands Univ Amsterdam Acad Med Ctr Dept Endocrinol & Metab NL-1105 AZ Amsterdam Netherlands Univ Rotterdam Hosp Dr Daniel den Hoed Canc Ctr Dept Surg Oncol Rotterdam Netherlands St Joseph Hosp Dept Internal Med Veldhoven Netherlands

Cytokines have been implicated in the pathogenesis of the euthyroid sick syndrome. Isolated limb perfusion (ILP) with recombinant human tumor necrosis factor alpha (rTNF) and melphalan in patients with melanoma or sarcoma is accompanied by high systemic TNF: levels. We examined the prolonged effects (7 days) of ILP on thyroid hormone metabolism with respect to induction and recovery of the euthyroid sick syndrome in six cancer patients. After ILP, when the limb is reconnected to the systemic circulation, leakage of residual rTNF resulted in systemic peak levels at. 10 minutes postperfusion followed by a parallel increase in plasma interleukin-6 (IL-6) and cortisol, with maximum levels at 4 hours (P < .05). A rapid decrease was observed at 5 minutes for plasma triiodothyronine (T3), reverse T3 (rT3), thyroxine (T4), and thyroxine-binding globulin (TBG) (P < .05), whereas free T4 (FT4) and T3-uptake showed a sharp increase, with peak levels at 5 minutes (P < .05). T3, T4, and TBG levels remained low until 24 hours after ILP. In contrast, rT3 increased above pretreatment values to maximum levels at 24 hours (P < .05), Plasma thyrotropin (TSH) showed an initial decrease at 4 hours postperfusion (P < .05) but exceeded pretreatment values from day 1 to day 7 (by +94% +/- 43% to +155% +/- 66%, P < .05), preceding the recovery of T4 and T3 levels. T3 and rT3 returned to initial values at day 4. T4 and TBG levels recovered at day 2. T4 exceeded basal values at days 5 to 7 (P < .05). It is concluded that ILP with rTNF induces a euthyroid sick syndrome either directly or indirectly through other mediators such as IL-6 or cortisol. The recovery from this euthyroid sick syndrome is, at least in part, TSH-dependent, since the prolonged elevation of TSH values preceded and persisted during the normalization of T3 and the elevation of T4 levels. This biphasic pattern of induction of and recovery from the euthyroid sick syndrome may be a general feature of nonthyroidal disease. The

关键词： Euthyroid Sick Syndromes Melanoma Sarcoma Soft Tissue Neoplasms Tumor Necrosis Factor 功能正常甲状腺病综合征黑色素瘤肉瘤软组织肿瘤肿瘤坏死因子 Thyroxine inductive logic programming Melanoma Euthyroid Sick Syndromes Sarcoma Tumor Necrosis Factor-alpha Thyroxine-Binding Globulin Soft Tissue Neoplasms Triiodothyronine, Reverse Tumor Necrosis Factors

来源：评论

学校读者我要写书评

暂无评论

A Combined Analytical and Search-Based Approach for the inductive Synthesis of Functional Programs

引用

KUNSTLICHE INTELLIGENZ 2011年第2期25卷 179-182页

作者： Kitzelmann, Emanuel Int Comp Sci Inst 1947 Ctr StSuite 600 Berkeley CA 94704 USA

inductive program synthesis addresses the problem of automatically generating (declarative) recursive programs from ambiguous specifications such as input/output examples. Potential applications range from software development to intelligent agents that learn in recursive domains. Current systems suffer from either strong restrictions regarding the form of inducible programs or from blind search in vast program spaces. The main contribution of my dissertation (Kitzelmann, Ph.D. thesis, 2010) is the algorithm IGOR2 for the induction of functional programs. It is based on search in program spaces but derives candidate programs directly from examples, rather than using them as test cases, and thereby prunes many programs. Experiments show promising results.

关键词： Background Knowledge Functional Program inductive logic programming Intelligent Agent Program Space

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：