Integrating Learning and Reasoning with Deep Logic Models

Marra, Giuseppe; Giannini, Francesco; Diligenti, Michelangelo; Gori, Marco

doi:10.1007/978-3-030-46147-8_31

Cited by 37 publications

(37 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…to constrain it. The imposing of particular formal, highlevel, conceptual rules on AI models is an ongoing topic of research in many machine learning domains (Hu et al, 2016;Marra et al, 2020). In the context of AMT, such rules could comprise the possible note durations or interval sizes to be used, which should be parameters accessible to the user of an AMT software.…”

Section: Discussionmentioning

confidence: 99%

Humanities and engineering perspectives on music transcription

Widdess

Holzapfel

Benetos

et al. 2021

Digital Scholarship in the Humanities

View full text Add to dashboard Cite

Music transcription is a process of creating a notation of musical sounds. It has been used as a basis for the analysis of music from a wide variety of cultures. Recent decades have seen an increasing amount of engineering research within the field of Music Information Retrieval that aims at automatically obtaining music transcriptions in Western staff notation. However, such approaches are not widely applied in research in ethnomusicology. This article aims to bridge interdisciplinary gaps by identifying aspects of proximity and divergence between the two fields. As part of our study, we collected manual transcriptions of traditional dance tune recordings by eighteen transcribers. Our method employs a combination of expert and computational evaluation of these transcriptions. This enables us to investigate the limitations of automatic music transcription (AMT) methods and computational transcription metrics that have been proposed for their evaluation. Based on these findings, we discuss promising avenues to make AMT more useful for studies in the Humanities. These are, first, assessing the quality of a transcription based on an analytic purpose; secondly, developing AMT approaches that are able to learn conventions concerning the transcription of a specific style; thirdly, a focus on novice transcribers as users of AMT systems; and, finally, considering target notation systems different from Western staff notation.

show abstract

Section: Discussionmentioning

confidence: 99%

Humanities and engineering perspectives on music transcription

Widdess

Holzapfel

Benetos

et al. 2021

Digital Scholarship in the Humanities

View full text Add to dashboard Cite

show abstract

“…In contrast, DRAIL focuses on learning multiple interdependent decisions from data, handling and requiring supervision for all unknown atoms in a given example. Lastly, Deep Logic Models (DLMs) (Marra et al, 2019) learn a set of parameters to encode atoms in a probabilistic logic program. Similarly to Donadello et al (2017) Deep structured models.…”

Section: Deep Classifiers and Probabilistic Inferencementioning

confidence: 99%

Modeling Content and Context with Deep Relational Learning

Pacheco

Goldwasser

2021

Transactions of the Association for Computational Linguistics

View full text Add to dashboard Cite

Building models for realistic natural language tasks requires dealing with long texts and accounting for complicated structural dependencies. Neural-symbolic representations have emerged as a way to combine the reasoning capabilities of symbolic methods, with the expressiveness of neural networks. However, most of the existing frameworks for combining neural and symbolic representations have been designed for classic relational learning tasks that work over a universe of symbolic entities and relations. In this paper, we present DRaiL, an open-source declarative framework for specifying deep relational models, designed to support a variety of NLP scenarios. Our framework supports easy integration with expressive language encoders, and provides an interface to study the interactions between representation, inference and learning.

show abstract

“…However this approach does not allow any learning in the aspect of the logic rules, since they remain constant during training. In another work, G. Marra et al [12], integrate first order logic rules with deep learning in both training and evaluation settings. The approach of G. Marra et al allows the learning of the weights of the neural network and the parameters of the integrated logic rules that are used for reasoning.…”

Section: Related Workmentioning

confidence: 99%

Logic Rules Meet Deep Learning: A Novel Approach for Ship Type Classification

Pitsikalis¹,

Lisitsa²,

Si³

2021

Preprint

View full text Add to dashboard Cite

The shipping industry is an important component of the global trade and economy, however in order to ensure law compliance and safety it needs to be monitored. In this paper, we present a novel Ship Type classification model that combines vessel transmitted data from the Automatic Identification System, with vessel imagery. The main components of our approach are the Faster R-CNN Deep Neural Network and a Neuro-Fuzzy system with IF-THEN rules. We evaluate our model using real world data and showcase the advantages of this combination while also compare it with other methods. Results show that our model can increase prediction scores by up to 15.4% when compared with the next best model we considered, while also maintaining a level of explainability as opposed to common black box approaches.

show abstract

Integrating Learning and Reasoning with Deep Logic Models

Cited by 37 publications

References 15 publications

Humanities and engineering perspectives on music transcription

Humanities and engineering perspectives on music transcription

Modeling Content and Context with Deep Relational Learning

Logic Rules Meet Deep Learning: A Novel Approach for Ship Type Classification

Contact Info

Product

Resources

About