Inductive logic programming at 30

Cropper, Andrew; Dumančić, Sebastijan; Evans, Rhett; Muggleton, Stephen

doi:10.1007/s10994-021-06089-1

Cited by 30 publications

(18 citation statements)

References 85 publications

(189 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…• Huang et al [92] Noisy Label Optimization [131] One-vs-All Classification [96] Domain Similarity & Prediction Shift [42] Feature-Label Matching [40] Bi-Level Optimization [36,49,93,136] Distance-based Re-weighting [87] Distribution Divergence Minimization [15] Adversarial Feature Adaptation [42] Feature Adaptation Semi-Supervised OOD Detection Detection-Specific Metric Learning [138] Fig. 5.…”

Section: Label Distribution Mismatchmentioning

confidence: 99%

“…Therefore, how to exploit both unlabeled data and symbolic domain knowledge in machine learning is a worth-studied problem. There are some efforts have been devoted to utilizing domain knowledge, such as inductive logic programming [138], statistical relational learning [139], and probabilistic logic programming [140]. Recently, a novel framework Abductive learning [141], [142] is proposed to bridges machine learning and symbolic reasoning in tasks with raw input space and symbolic domain knowledge.…”

Section: Robust Ssl With Domain Knowledgementioning

confidence: 99%

See 1 more Smart Citation

Robust Deep Semi-Supervised Learning: A Brief Introduction

Guo¹,

Zhou²,

Li³

2022

Preprint

View full text Add to dashboard Cite

Semi-supervised learning (SSL) is the branch of machine learning that aims to improve learning performance by leveraging unlabeled data when labels are insufficient. Recently, SSL with deep models has proven to be successful on standard benchmark tasks. However, they are still vulnerable to various robustness threats in real-world applications as these benchmarks provide perfect unlabeled data, while in realistic scenarios, unlabeled data could be corrupted. Many researchers have pointed out that after exploiting corrupted unlabeled data, SSL suffers severe performance degradation problems. Thus, there is an urgent need to develop SSL algorithms that could work robustly with corrupted unlabeled data. To fully understand robust SSL, we conduct a survey study. We first clarify a formal definition of robust SSL from the perspective of machine learning. Then, we classify the robustness threats into three categories: i) distribution corruption, i.e., unlabeled data distribution is mismatched with labeled data; ii) feature corruption, i.e., the features of unlabeled examples are adversarially attacked; and iii) label corruption, i.e., the label distribution of unlabeled data is imbalanced. Under this unified taxonomy, we provide a thorough review and discussion of recent works that focus on these issues. Finally, we propose possible promising directions within robust SSL to provide insights for future research.

show abstract

Section: Label Distribution Mismatchmentioning

confidence: 99%

Section: Robust Ssl With Domain Knowledgementioning

confidence: 99%

Robust Deep Semi-Supervised Learning: A Brief Introduction

Guo¹,

Zhou²,

Li³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Background knowledge is used together with examples to induce a hypothesis, in the form of a logic program, describing positive and negative examples in given data. ILP is applied to, e.g., scientific discovery, robotics, program analysis (Cropper et al, 2022), with neurosymbolic approaches to machine vision problems (Dai et al, 2015;Varghese et al, 2021).…”

Section: Related Workmentioning

confidence: 99%

An Algebraic Approach to Learning and Grounding

Björklund¹,

Lindström²,

Drewes³

2022

Preprint

View full text Add to dashboard Cite

We consider the problem of learning the semantics of composite algebraic expressions from examples. The outcome is a versatile framework for studying learning tasks that can be put into the following abstract form: The input is a partial algebra A and a finite set of samples (ϕ 1 , O 1 ), (ϕ 2 , O 2 ), . . ., each consisting of an algebraic term ϕ i and a set of objects O i . The objective is to simultaneously fill in the missing algebraic operations in A and ground the variables of every ϕ i in O i , so that the combined value of the terms is optimised. We demonstrate the applicability of this framework through case studies in grammatical inference, picture-language learning, and the grounding of logic scene descriptions.

show abstract

“…Learning recursive programs has long been considered a difficult problem for ILP [81,17]. The power of recursion is that an infinite number of computations can be described by a finite recursive program [107].…”

Section: Recursionmentioning

confidence: 99%