Structure Learning via Parameter Learning

Wang, William Yang; Mazaitis, Kathryn; Cohen, William W.

doi:10.1145/2661829.2662022

Cited by 32 publications

(20 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Metarules (Cropper and Tourret 2020) are another popular syntactic bias used by many ILP approaches (De Raedt and Bruynooghe 1992;Wang et al 2014;Albarghouthi et al 2017;Kaminski et al 2018), including Metagol (Muggleton et al 2015;Cropper and Muggleton 2016) and, to an extent 5 , ∂ILP (Evans and Grefenstette 2018). A metarule is a higher-order clause which defines the exact form of clauses in the hypothesis space.…”

Section: Language Biasmentioning

confidence: 99%

Learning programs by learning from failures

2021

View full text Add to dashboard Cite

We describe an inductive logic programming (ILP) approach called learning from failures. In this approach, an ILP system (the learner) decomposes the learning problem into three separate stages: generate, test, and constrain. In the generate stage, the learner generates a hypothesis (a logic program) that satisfies a set of hypothesis constraints (constraints on the syntactic form of hypotheses). In the test stage, the learner tests the hypothesis against training examples. A hypothesis fails when it does not entail all the positive examples or entails a negative example. If a hypothesis fails, then, in the constrain stage, the learner learns constraints from the failed hypothesis to prune the hypothesis space, i.e. to constrain subsequent hypothesis generation. For instance, if a hypothesis is too general (entails a negative example), the constraints prune generalisations of the hypothesis. If a hypothesis is too specific (does not entail all the positive examples), the constraints prune specialisations of the hypothesis. This loop repeats until either (i) the learner finds a hypothesis that entails all the positive and none of the negative examples, or (ii) there are no more hypotheses to test. We introduce Popper, an ILP system that implements this approach by combining answer set programming and Prolog. Popper supports infinite problem domains, reasoning about lists and numbers, learning textually minimal programs, and learning recursive programs. Our experimental results on three domains (toy game problems, robot strategies, and list transformations) show that (i) constraints drastically improve learning performance, and (ii) Popper can outperform existing ILP systems, both in terms of predictive accuracies and learning times.

show abstract

Section: Language Biasmentioning

confidence: 99%

Learning programs by learning from failures

2021

View full text Add to dashboard Cite

show abstract

“…Recently, researchers (Rocktäschel & Riedel, 2017) have proposed a "differentiable theorem prover", in which a proof for an example is unrolled into a network. Their system includes representation-learning as a component, as well as a template-instantiation approach like that of (Wang, Mazaitis, & Cohen, 2014), allowing structure learning. However, unlike the case for TensorLog, the proof procedure can produce very large proof trees, leading to proofs (and neural networks) that are of size exponential in the KB.…”

Section: Hybrid Logical/neural Systemsmentioning

confidence: 99%

“…12 . A second was learning the most common relation, "affects", from UMLS, using a rule set learned by the algorithm of (Wang et al, 2014). Also, motivated by comparisons between ProPPR and embedding-based approaches to knowledge-base completion (Wang & Cohen, 2016), we compared to ProPPR on two relation-prediction tasks involving WordNet, using rules from the non-recursive theories used in (Wang & Cohen, 2016).…”

Section: Small Relational Learning Tasksmentioning

confidence: 99%

TensorLog: A Probabilistic Database Implemented Using Deep-Learning Infrastructure

Cohen¹,

Yang²,

Mazaitis³

2020

jair

View full text Add to dashboard Cite

We present an implementation of a probabilistic first-order logic called TensorLog, in which classes of logical queries are compiled into differentiable functions in a neural-network infrastructure such as Tensorflow or Theano. This leads to a close integration of probabilistic logical reasoning with deep-learning infrastructure: in particular, it enables high-performance deep learning frameworks to be used for tuning the parameters of a probabilistic logic. The integration with these frameworks enables use of GPU-based parallel processors for inference and learning, making TensorLog the first highly parallellizable probabilistic logic. Experimental results show that TensorLog scales to problems involving hundreds of thousands of knowledge-base triples and tens of thousands of examples.

show abstract

“…Early and more recent work on learning structures [25] and use of background knowledge included advanced in Inductive Logic Programming (ILP) and hierarchical learning methods. In the ILP, the background knowledge is defined as a set of relations (predicates) that can be used in the definition of the target concept [26].…”

Section: Supervised Learningmentioning

confidence: 99%

Guiding Supervised Learning by Bio-Ontologies in Medical Data Analysis

Wojtusiak

Min

Elashkar

et al. 2018

IFIP Advances in Information and Communication Technology

View full text Add to dashboard Cite

Ontologies are popular way of representing knowledge and semantics of data in medical and health fields. Surprisingly, few machine learning methods allow for encoding semantics of data and even fewer allow for using ontologies to guide learning process. This paper discusses the use of data semantics and ontologies in health and medical applications of supervised learning, and particularly describes how the Unified Medical Language System (UMLS) is used within AQ21 rule learning software. Presented concepts are illustrated using two applications based on distinctly different types of data and methodological issues.

show abstract

Structure Learning via Parameter Learning

Cited by 32 publications

References 28 publications

Learning programs by learning from failures

Learning programs by learning from failures

TensorLog: A Probabilistic Database Implemented Using Deep-Learning Infrastructure

Guiding Supervised Learning by Bio-Ontologies in Medical Data Analysis

Contact Info

Product

Resources

About