52,939 Hits in 5.1 sec

Neural-Symbolic Learning and Reasoning: A Survey and Interpretation [article]

Tarek R. Besold, Artur d'Avila Garcez, Sebastian Bader, Howard Bowman, Pedro Domingos, Pascal Hitzler, Kai-Uwe Kuehnberger, Luis C. Lamb, Daniel Lowd, Priscila Machado Vieira Lima, Leo de Penning, Gadi Pinkas, Hoifung Poon (+1 others)
2017 arXiv   pre-print
This joint survey reviews the personal ideas and views of several researchers on neural-symbolic learning and reasoning.  ...  We then proceed to describe the realisations of neural-symbolic computation, systems, and applications. Finally we present the challenges facing the area and avenues for further research.  ...  The challenges for neural-symbolic integration today emerge from the goal of effective integration, expressive reasoning, and robust learning.  ... 
arXiv:1711.03902v1 fatcat:3fod6z4oevhplpv2hzlkguyqiu

Neural Semirings

Pedro Zuidberg Dos Martires
2021 International Workshop on Neural-Symbolic Learning and Reasoning  
In this paper we relax this assumption and render the reasoning operators (the semiring operations) subject to learning by replacing predefined operations with learnable neural networks.  ...  This unlocks the learn to reason paradigm in the semiring reasoning setting.  ...  Contribution Learning within the semiring reasoning framework of algebraic model counting has mainly been concerned with learning a mapping from literals to elements of a given semiring.  ... 
dblp:conf/nesy/Martires21 fatcat:jmkwhvi2b5he5lhvjx3utyxo6q

Learning Arithmetic from Handwritten Images with the Aid of Symbols

Daniel L. Silver, Ahmed Galila
2021 International Workshop on Neural-Symbolic Learning and Reasoning  
The networks are trained in the presence and absence of symbols for the noisy digits.  ...  This is accomplished by learning to map the noisy examples of a concept to a the correct class label or symbol.  ...  Deep recurrent neural networks (RNNs) experience similar challenges when learning a new algorithm from a small dataset containing examples of input-output sequences.  ... 
dblp:conf/nesy/SilverG21 fatcat:ekvwgfpomvhh7kxsqkr4qz423e

Composing Neural Learning and Symbolic Reasoning with an Application to Visual Discrimination [article]

Adithya Murali, Atharva Sehgal, Paul Krogmeier, P. Madhusudan
2022 arXiv   pre-print
We propose a compositional neurosymbolic framework that combines a neural network to detect objects and relationships with a symbolic learner that finds interpretable discriminators.  ...  We create large classes of VDP datasets involving natural and artificial images and show that our neurosymbolic framework performs favorably compared to several purely neural approaches.  ...  Acknowledgments This work is supported in part by a research grant from Amazon and a Discovery Partner's Institute (DPI) science team seed grant.  ... 
arXiv:1907.05878v2 fatcat:ch4upg2s7bfftbx7fy6ntw4fta

Unifying Neural Learning and Symbolic Reasoning for Spinal Medical Report Generation [article]

Zhongyi Han, Benzheng Wei, Yilong Yin, Shuo Li
2020 arXiv   pre-print
In this paper, we propose the neural-symbolic learning (NSL) framework that performs human-like learning by unifying deep neural learning and symbolic logical reasoning for the spinal medical report generation  ...  However, it is incredibly challenging because it is an extremely complicated task that involves visual perception and high-level reasoning processes.  ...  NSL combines neural learning and symbolic reasoning in a mutually beneficial way.  ... 
arXiv:2004.13577v1 fatcat:oh5aka5zr5be3ipd7qqnyhikzy

Neural-Symbolic Argumentation Mining: An Argument in Favor of Deep Learning and Reasoning

Andrea Galassi, Kristian Kersting, Marco Lippi, Xiaoting Shao, Paolo Torroni
2020 Frontiers in Big Data  
In this position paper, we posit that neural-symbolic and statistical relational learning could play a crucial role in the integration of symbolic and sub-symbolic methods to achieve this goal.  ...  Deep learning is bringing remarkable contributions to the field of argumentation mining, but the existing approaches still need to fill the gap toward performing advanced reasoning tasks.  ...  ACKNOWLEDGMENTS We thank the reviewers for their comments and contributions, which have increased the quality of this work.  ... 
doi:10.3389/fdata.2019.00052 pmid:33693375 pmcid:PMC7931943 fatcat:mtp5xigtlndwvnz7olwcpvd6na

Neural-Symbolic Computing: An Effective Methodology for Principled Integration of Machine Learning and Reasoning [article]

Artur d'Avila Garcez, Marco Gori, Luis C. Lamb, Luciano Serafini, Michael Spranger, Son N. Tran
2019 arXiv   pre-print
Neural-symbolic computing has been an active topic of research for many years, reconciling the advantages of robust learning in neural networks and reasoning and interpretability of symbolic representation  ...  In this paper, we survey recent accomplishments of neural-symbolic computing as a principled methodology for integrated machine learning and reasoning.  ...  reasoning and neural learning.  ... 
arXiv:1905.06088v1 fatcat:gm4f3ncukrbevpd7nq5yr75ar4

Extracting Argumentative Dialogues from the Neural Network that Computes the Dungean Argumentation Semantics

Yoshiaki Goto, Wataru Makiguchi, Hajime Sawamura
2011 International Workshop on Neural-Symbolic Learning and Reasoning  
We deal with the question how various argumentation semantics can have dialectical proof theories, and describe a possible answer to it by extracting or generating symbolic dialogues from the neural network  ...  In this paper, we are concerned with the opposite direction from neural network computation to symbolic argumentation/dialogue.  ...  programming and non-monotonic reasoning.  ... 
dblp:conf/nesy/GotoMS11 fatcat:bsxaexkwubebxpgglewbfvqh3u

Explainable and Interpretable Diabetic Retinopathy Classification Based on Neural-Symbolic Learning [article]

Se-In Jang, Michael J.A. Girard, Alexandre H. Thiery
2022 arXiv   pre-print
In this paper, we propose an explainable and interpretable diabetic retinopathy (ExplainDR) classification model based on neural-symbolic learning.  ...  To gain explainability, a highlevel symbolic representation should be considered in decision making.  ...  Neural-Symbolic Learning The goal of neural-symbolic learning is to provide a coherent, unifying view for logic and connectionism to contribute to the modelling and understanding of cognition and, thereby  ... 
arXiv:2204.00624v1 fatcat:cnqmzghsizdpzepeuq4wamuuqi

Machine learning for information retrieval: Neural networks, symbolic learning, and genetic algorithms

Hsinchun Chen
1995 Journal of the American Society for Information Science  
More recently, information science researchers have turned to other newer artificial-intelligence-based inductive learning techniques including neural networks, symbolic learning, and genetic algorithms  ...  In the 198Os, knowledge-based techniques also made an impressive contribution to "intelligent" information retrieval and indexing.  ...  Acknowledgments This project was supported mainly by NSF Grant #IRI-92 114 18, 1992-1994 (NSF/CISE, Division of Information, Robotics, and Intelligent Systems).  ... 
doi:10.1002/(sici)1097-4571(199504)46:3<194::aid-asi4>;2-s fatcat:4kkxuco74fe2vgbtnqax7eucim

Representation Learning on Visual-Symbolic Graphs for Video Understanding [article]

Effrosyni Mavroudi, Benjamín Béjar Haro, René Vidal
2020 arXiv   pre-print
; and d) performs global reasoning in the semantic space.  ...  types of interactions, and (2) a symbolic graph that models semantic relationships.  ...  Experiments To demonstrate the effectiveness and generality of our method, we conduct experiments on three challenging video understanding tasks that require reasoning about interactions between semantic  ... 
arXiv:1905.07385v2 fatcat:6fz7xtbmhvh5hms2g5rplcghlm

Planning chemical syntheses with deep neural networks and symbolic AI

Marwin H. S. Segler, Mike Preuss, Mark P. Waller
2018 Nature  
These deep neural networks were trained on 12 million reactions, which represents essentially all reactions ever published in organic chemistry.  ...  Our system solves almost twice as many molecules and is 30 times faster in comparison to the traditional search method based on extracted rules and hand-coded heuristics.  ...  Discussion We have shown that MCTS combined with deep neural networks and symbolic rules can be used effectively to perform chemical synthesis planning.  ... 
doi:10.1038/nature25978 pmid:29595767 fatcat:yb6scvw5ffanfl446ho3nrbcoi

Top-Down and Bottom-Up Interactions between Low-Level Reactive Control and Symbolic Rule Learning in Embodied Agents

Clément Moulin-Frier, Xerxes D. Arsiwalla, Jordi-Ysard Puigbo, Martí Sánchez-Fibla, Armin Duff, Paul F. M. J. Verschure
2016 Neural Information Processing Systems  
symbolic rule learning in embodied agents.  ...  The interaction of these modules in a closed-loop fashion suggests how symbolic representations might have been shaped from low-level behaviors and recruited for behavior optimization.  ...  Top-down and bottom-up approaches thus reflect different aspects of cognition: high-level symbolic reasoning for the former and low-level embodied behaviors for the latter.  ... 
dblp:conf/nips/Moulin-FrierAPS16 fatcat:zepcxsghffhcvgiigrowskpfee

Weakly Supervised Neural Symbolic Learning for Cognitive Tasks

Jidong Tian, Yitian Li, Wenqing Chen, Liqiang Xiao, Hao He, Yaohui Jin
In this paper, we propose WS-NeSyL, a weakly supervised neural symbolic learning model for cognitive tasks with logical reasoning.  ...  However, it is challenging to apply NeSyL to these cognitive tasks because of the lack of supervision, the non-differentiable manner of the symbolic system, and the difficulty to probabilistically constrain  ...  Technology Commission (19JC1410102), the Shanghai Municipal Science and Technology Major Project (2021SHZDZX0102), the Shanghai Pujiang Program (21PJ1407300), and the Shanghai Science and Technology Innovation  ... 
doi:10.1609/aaai.v36i5.20533 fatcat:ir5gx4rv65fexiahwy4u4pei3u

Mining Multiple Models

Graham J. Williams
2006 Contributions to Probability and Statistics: Applications and Challenges  
The idea was first introduced through the concept of multiple inductive learning (MIL) (Williams, 1988 (Williams, , 1991 and further developed in practise as mining the data mine (Williams and Huang, 1997  ...  Many data mining advances that have since emerged have further developed the idea: multiple modelling, ensemble learning, bagging, and boosting all help the data miner explore different ideas and look  ...  Traditionally, this means building decision trees or logistic regression models or neural networks. Data underlies data mining and comes in many shapes and sizes.  ... 
doi:10.1142/9789812772466_0022 fatcat:k57vtscysvbjdmage7l73yl57u
« Previous Showing results 1 — 15 out of 52,939 results