Harnessing Deep Neural Networks with Logic Rules

Hu, Zhiting; Ma, Xuezhe; Liu, Zhengzhong; Hovy, Eduard; Xing, Eric P.

doi:10.48550/arxiv.1603.06318

Cited by 110 publications

(85 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As shown in [10], adaptation of logical knowledge as constraints during the learning process has generated promising results, that reinforces the attempts to use ontologies as background knowledge. The area of neuro-symbolic approaches also provides insights into the use of logical knowledge during the training of artificial neural networks [22].…”

Section: Related Worksupporting

confidence: 54%

See 1 more Smart Citation

Ontology-based n-ball Concept Embeddings Informing Few-shot Image Classification

Jayathilaka,

Mu,

Sattler

2021

Preprint

View full text Add to dashboard Cite

We propose a novel framework named ViOCE that integrates ontology-based background knowledge in the form of n-ball concept embeddings into a neural network based vision architecture. The approach consists of two components -converting symbolic knowledge of an ontology into continuous space by learning n-ball embeddings that capture properties of subsumption and disjointness, and guiding the training and inference of a vision model using the learnt embeddings. We evaluate ViOCE using the task of few-shot image classification, where it demonstrates superior performance on two standard benchmarks.

show abstract

Section: Related Worksupporting

confidence: 54%

“…Out of these, the use of scene graphs, probabilistic ontologies and firstorder logic rules grab the attention as promising paths to explore. Investigations into the use of background knowledge in the form of first-Order Logic (FOL) is prominently seen in several studies [10].…”

Section: Related Workmentioning

confidence: 99%

Ontology-based n-ball Concept Embeddings Informing Few-shot Image Classification

Jayathilaka,

Mu,

Sattler

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Domain-specific knowledge of correctness of detection and recognition can be utilized to mitigate requirement of hand labelled data. Parallely there has been related work on using domain knowledge to regularize model posteriors [3,4]. We formalize domain knowledge of correctness of recognized team names, time, or half/quarter as first-order logical rules and refer to them as Knowledge Constraints (KC).…”

Section: Approachmentioning

confidence: 99%

Distantly Supervised Semantic Text Detection and Recognition for Broadcast Sports Videos Understanding

Shah

Biswas

Ramadoss

et al. 2021

Proceedings of the 29th ACM International Conference on Multimedia

View full text Add to dashboard Cite

Comprehensive understanding of key players and actions in multiplayer sports broadcast videos is a challenging problem. Unlike in news or finance videos, sports videos have limited text. While both action recognition for multiplayer sports and detection of players has seen robust research, understanding contextual text in video frames still remains one of the most impactful avenues of sports video understanding. In this work we study extremely accurate semantic text detection and recognition in sports clocks, and challenges therein. We observe unique properties of sports clocks, which makes it hard to utilize general-purpose pre-trained detectors and recognizers, so that text can be accurately understood to the degree of being used to align to external knowledge. We propose a novel distant supervision technique to automatically build sports clock datasets. Along with suitable data augmentations, combined with any state-of-the-art text detection and recognition model architectures, we extract extremely accurate semantic text. Finally, we share our computational architecture pipeline to scale this system in industrial setting and proposed a robust dataset for the same to validate our results. CCS CONCEPTS• Computing methodologies → Activity recognition and understanding; Visual content-based indexing and retrieval; Scene understanding; Object detection; Object recognition.

show abstract

“…In the context of neural-based dialogue systems, this line is pursued by using constrained rules (Jhunjhunwala et al, 2020), logical rules to be used in inductive logic programming (Zhou et al, 2020) or declarative language (Altszyler et al, 2020). These rules and models can be easily included in the existing dialogue state tracking models to guide the training and predictions phases without additional learning parameters (Hu et al, 2016;van Krieken et al, 2020). These models obtain the same advantage of the user simulator and in addition overcome the problem of the evaluation of the user-simulator itself.…”

Section: Background and Related Workmentioning

confidence: 99%

Every time I fire a conversational designer, the performance of the dialog system goes down

Xompero¹,

Mastromattei²,

Salman³

et al. 2021

Preprint

View full text Add to dashboard Cite

Incorporating explicit domain knowledge into neural-based task-oriented dialogue systems is an effective way to reduce the need of large sets of annotated dialogues. In this paper, we investigate how the use of explicit domain knowledge of conversational designers affects the performance of neural-based dialogue systems. To support this investigation, we propose the Conversational-Logic-Injection-in-Neural-Network system (CLINN) where explicit knowledge is coded in semilogical rules. By using CLINN, we evaluated semi-logical rules produced by a team of differently-skilled conversational designers. We experimented with the Restaurant topic of the MultiWOZ dataset. Results show that external knowledge is extremely important for reducing the need of annotated examples for conversational systems. In fact, rules from conversational designers used in CLINN significantly outperform a state-of-the-art neural-based dialogue system.

show abstract

Harnessing Deep Neural Networks with Logic Rules

Cited by 110 publications

References 22 publications

Ontology-based n-ball Concept Embeddings Informing Few-shot Image Classification

Ontology-based n-ball Concept Embeddings Informing Few-shot Image Classification

Distantly Supervised Semantic Text Detection and Recognition for Broadcast Sports Videos Understanding

Every time I fire a conversational designer, the performance of the dialog system goes down

Contact Info

Product

Resources

About