This study presents a review of applications of inductive logic programming (ILP) for robotic systems. The aim of the paper is to demonstrate the different methods of applying ILP to a robotic system and to also highl...
详细信息
ISBN:
(数字)9783031492990
ISBN:
(纸本)9783031492983;9783031492990
This study presents a review of applications of inductive logic programming (ILP) for robotic systems. The aim of the paper is to demonstrate the different methods of applying ILP to a robotic system and to also highlight some of the limitations that already exist. ILP can aid in the development of explainable and trustworthy robotics systems.
An interesting feature that traditional approaches to inductive logic programming are missing is the ability to treat noisy and non-logical data. Neural-symbolic approaches to inductive logic programming have been rec...
详细信息
ISBN:
(纸本)9783031157073;9783031157066
An interesting feature that traditional approaches to inductive logic programming are missing is the ability to treat noisy and non-logical data. Neural-symbolic approaches to inductive logic programming have been recently proposed to combine the advantages of inductive logic programming, in terms of interpretability and generalization capability, with the characteristic capacity of deep learning to treat noisy and nonlogical data. This paper concisely surveys and briefly compares three promising neural-symbolic approaches to inductive logic programming that have been proposed in the last five years. The considered approaches use Datalog dialects to represent background knowledge, and they are capable of producing reusable logical rules from noisy and non-logical data. Therefore, they provide an effective means to combine logical reasoning with state-of-the-art machine learning.
We present a relational multi-agent reinforcement learning algorithm in which two agents work together to achieve a goal in an environment represented by structured entities and relations. Our proposal takes a hybrid ...
详细信息
ISBN:
(数字)9783031159312
ISBN:
(纸本)9783031159312;9783031159305
We present a relational multi-agent reinforcement learning algorithm in which two agents work together to achieve a goal in an environment represented by structured entities and relations. Our proposal takes a hybrid connectionist-symbolic approach, where a classical actorcritic method with an iterative weight update scheme is used to guide the derivation of an agent's policy, which is purely expressed as first-order logic. A recent technique, differentiable inductive logic programming, is applied to integrate these two parts into a trainable system. We tailor the centralized training with decentralized execution framework to meet the symbolic-represented underlying structure. Agents are designed to communicate with one another in terms of logical predicates to alleviate the partially observable problem prevalent in the multi-agent setting. Empirical studies on the classical grid-world task demonstrate that the proposed method can learn close to optimal strategies and has better interpretability than traditional reinforcement learning approaches.
Despite the impressive performance of Deep Neural Networks (DNNs), they usually lack the explanatory power of disciplines such as logicprogramming. Even though they can learn to solve very difficult problems, the lea...
详细信息
Despite the impressive performance of Deep Neural Networks (DNNs), they usually lack the explanatory power of disciplines such as logicprogramming. Even though they can learn to solve very difficult problems, the learning is usually implicit and it is very difficult, if not impossible, to interpret and decipher the underlying explanations that is implicitly stored in the weights of the neural network models. On the other hand, standard logicprogramming is usually limited in scope and application compared to the DNNs. The objective of this dissertation is to bridge the gap between these two disciplines by presenting a novel paradigm for learning algorithmic and discrete tasks via neural networks. This novel approach, uses the differentiable neural network to design interpretable and explanatory models that can learn and represent Boolean functions efficiently. We will investigate the application of these differentiable Neural logic (dNL) networks in disciplines such as inductive logic programming, Relational Reinforcement Learning, as well as in discrete algorithmic tasks such as decoding LDPC codes over Binary erasure Channels. inductive logic programming (ILP) is an important branch of machine learning that uses formal logic to define and solve problems. Compared to the DNNs, ILP provides high degree of interpretability and can learn from small number of examples and has superior generalization ability. However, standard ILP solvers are not differentiable and cannot benefit from the impressive power of DNNs. In this dissertation, we reformulate the ILP as a differentiable neural network by exploiting the explanatory power of dNL networks. This novel neural based ILP solver (dNL-ILP) is capable of learning auxiliary and invented predicates as well as learning complex recursive predicates. We show that dNL-ILP outperforms the current state of the art ILP solvers in a variety of benchmark algorithmic tasks as well as larger scale relational classification tasks. In
This work presents some approaches to overcome current Reinforcement Learning limits. We implement a simple virtual environment and some state-of-the-art Reinforcement Learning algorithms for testing and producing a b...
详细信息
ISBN:
(纸本)9783030456900;9783030456917
This work presents some approaches to overcome current Reinforcement Learning limits. We implement a simple virtual environment and some state-of-the-art Reinforcement Learning algorithms for testing and producing a baseline for comparison. Then we implement a Relational Reinforcement Learning algorithm that shows superior performance to the baseline but requires introducing human knowledge. We also propose that Model-based Reinforcement Learning can help us overcome some of the barriers. For better World models, we explore inductive logic programming methods, such as First-Order inductive Learner, and develop an improved version of it, more adequate to Reinforcement Learning environments. Finally we develop a novel Neural Network architecture, the inductivelogic Neural Network, to fill the gaps of the previous implementations, that shows great promise.
Ontologies are key concepts in the semantic web and have an impressive role which comprise the biggest and the most prominent part of the infrastructure in this realm of web research. By fast growth of the semantic we...
详细信息
Ontologies are key concepts in the semantic web and have an impressive role which comprise the biggest and the most prominent part of the infrastructure in this realm of web research. By fast growth of the semantic web and also, the variety of its applications, ontology mapping (ontology alignment) has been transformed into a crucial issue in the realm of computer science. Several approaches are introduced for ontology alignment during these last years, but developing more accurate and efficient algorithms and finding new effective techniques and algorithms for this problem is an interesting research area since real-world applications with respect to their more complicated concepts need more efficient algorithms. In this paper, we illustrated a new ontology mapping method based on learning using inductive logic programming (ILP), and show how the ILP can be used to solve the ontology mapping problem. As a matter of fact, in this approach, an ontology which is described in OWL format is interpreted to first-order logic. Then, with the use of learning based on inductivelogic, the existing hidden rules and relationships between concepts are discovered and presented. Since the inductivelogic has high flexibility in solving problems such as discovering relationships between concepts and links, it also can be performed effectively in solving the ontology alignment problem. Our experimental results show that this technique yield to more accurate results comparing to other matching algorithms and systems, achieving an F-measure of 95.6% and 91% on two well-known reference datasets the Anatomy and the Library, respectively. (C) 2019 Elsevier Ltd. All rights reserved.
inductive logic programming (ILP) is a hot research field in machine learning. Although ILP has obtained great success in many domains, in most ILP system, deterministic search are used to search the hypotheses space,...
详细信息
inductive logic programming (ILP) is a hot research field in machine learning. Although ILP has obtained great success in many domains, in most ILP system, deterministic search are used to search the hypotheses space, and they are easy to trap in local optima. To overcome the shortcomings, an ILP system based on artificial bee colony (ABCILP) is proposed in this article. ABCILP adopts an ABC stochastic search to examine the hypotheses space, the shortcoming of deterministic search is conquered by stochastic search. ABCILP regard each first-order rule as a food source and propose some discrete operations to generate the neighborhood food sources. A new fitness is proposed and an adaptive strategy is adopted to determine the parameter of the new fitness. Experimental results show that: 1) the proposed new fitness function can more precisely measure the quality of hypothesis and can avoid generating an over-specific rule;2) the performance of ABCILP is better than other systems compared with it.
Modern search techniques either cannot efficiently incorporate human feedback to refine search results or cannot express structural or semantic properties of desired code. The key insight of our interactive code searc...
详细信息
ISBN:
(纸本)9781728108698
Modern search techniques either cannot efficiently incorporate human feedback to refine search results or cannot express structural or semantic properties of desired code. The key insight of our interactive code search technique ALICE is that user feedback can be actively incorporated to allow users to easily express and refine search queries. We design a query language to model the structure and semantics of code as logic facts. Given a code example with user annotations, ALICE automatically extracts a logic query from code features that are tagged as important. Users can refine the search query by labeling one or more examples as desired (positive) or irrelevant (negative). ALICE then infers a new logic query that separates positive examples from negative examples via active inductive logic programming. Our comprehensive simulation experiment shows that ALICE removes a large number of false positives quickly by actively incorporating user feedback. Its search algorithm is also robust to user labeling mistakes. Our choice of leveraging both positive and negative examples and using nested program structure as an inductive bias is effective in refining search queries. Compared with an existing interactive code search technique, ALICE does not require a user to manually construct a search pattern and yet achieves comparable precision and recall with much fewer search iterations. A case study with real developers shows that ALICE is easy to use and helps express complex code patterns.
We describe the Inspire system which participated in the first competition on inductive logic programming (ILP). Inspire is based on answer set programming (ASP). The distinguishing feature of Inspire is an ASP encodi...
详细信息
We describe the Inspire system which participated in the first competition on inductive logic programming (ILP). Inspire is based on answer set programming (ASP). The distinguishing feature of Inspire is an ASP encoding for hypothesis space generation: given a set of facts representing the mode bias, and a set of cost configuration parameters, each answer set of this encoding represents a single rule that is considered for finding a hypothesis that entails the given examples. Compared with state-of-the-art methods that use the length of the rule body as a metric for rule complexity, our approach permits a much more fine-grained specification of the shape of hypothesis candidate rules. The Inspire system iteratively increases the rule cost limit and thereby increases the search space until it finds a suitable hypothesis. The system searches for a hypothesis that entails a single example at a time, utilizing an ASP encoding derived from the encoding used in XHAIL. We perform experiments with the development and test set of the ILP competition. For comparison we also adapted the ILASP system to process competition instances. Experimental results show that the cost parameters for the hypothesis search space are an important factor for finding hypotheses to competition instances within tight resource bounds.
Ontologies are one of the important and effective parts of semantic web which constitute the infrastructure and background knowledge of this realm of web science. Finding valid mappings as much as possible between the...
详细信息
ISBN:
(纸本)9781538653647
Ontologies are one of the important and effective parts of semantic web which constitute the infrastructure and background knowledge of this realm of web science. Finding valid mappings as much as possible between the concepts or entities of ontologies, especially for the large ones, is a prominent task to align those concepts together and finally merge and integrate their ontologies to make a general and global ontology that is smaller and more flexible in many applications of semantic web. This paper describes a new learning-based ontology mapping method in which inductive logic programming (ILP) is used to learn ontology mapping using information gathered from instances of each entity in order to make some correct and valid alignments between concepts of different ontologies.
暂无评论