Filters








41,835 Hits in 8.1 sec

A Logic-Driven Framework for Consistency of Neural Models [article]

Tao Li, Vivek Gupta, Maitrey Mehta, Vivek Srikumar
2019 arXiv   pre-print
We propose a learning framework for constraining models using logic rules to regularize them away from inconsistency.  ...  We instantiate our framework on natural language inference, where experiments show that enforcing invariants stated in logic can help make the predictions of neural models both accurate and consistent.  ...  Acknowledgements We thank members of the NLP group at the University of Utah for their valuable insights and sug-gestions, especially Mattia Medina Grespan for pointing out R-fuzzy logic and S-fuzzy logic  ... 
arXiv:1909.00126v4 fatcat:acltbisc5ndy5draq4x6q6qkci

A Logic-Driven Framework for Consistency of Neural Models

Tao Li, Vivek Gupta, Maitrey Mehta, Vivek Srikumar
2019 Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)  
We propose a learning framework for constraining models using logic rules to regularize them away from inconsistency.  ...  We instantiate our framework on natural language inference, where experiments show that enforcing invariants stated in logic can help make the predictions of neural models both accurate and consistent.  ...  Acknowledgements We thank members of the NLP group at the University of Utah for their valuable insights and suggestions; and reviewers for pointers to related works, corrections, and helpful comments.  ... 
doi:10.18653/v1/d19-1405 dblp:conf/emnlp/LiGMS19 fatcat:l3q532dlmnapvo4o6mg2vcfn5a

Survey on Applications of Neurosymbolic Artificial Intelligence [article]

Djallel Bouneffouf, Charu C. Aggarwal
2022 arXiv   pre-print
Specifically, we introduce a taxonomy of common Neurosymbolic applications and summarize the state-of-the-art for each of those domains.  ...  In recent years, the Neurosymbolic framework has attracted a lot of attention in various applications, from recommender systems and information retrieval to healthcare and finance.  ...  versus distributed: here we check how symbolic information is represented within the neural system, one dedicated neurons for a symbolic piece of knowledge) or a representation consists of a larger number  ... 
arXiv:2209.12618v1 fatcat:v37shgkahzdgznmob2pq4qimli

Abductive learning: towards bridging machine learning and logical reasoning

Zhi-Hua Zhou
2019 Science China Information Sciences  
leverage of perception and reasoning, where perception corresponds to a data-driven process that can be realized by machine learning whereas reasoning corresponds to a knowledge-driven process that can  ...  The exploration of abductive learning will possibly provide new approaches for developing a unified framework that accommodates learning and reasoning.  ... 
doi:10.1007/s11432-018-9801-4 fatcat:ae6rri4oybhtvno52abju6loda

Augmenting Neural Networks with First-order Logic

Tao Li, Vivek Srikumar
2019 Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics  
Our framework systematically compiles logical statements into computation graphs that augment a neural network without extra learnable parameters or manual redesign.  ...  In this paper, we present a novel framework for introducing declarative knowledge to neural network architectures in order to guide training and prediction.  ...  Acknowledgements We thank members of the NLP group at the University of Utah for their valuable insights and suggestions; and reviewers for pointers to related works, corrections, and helpful comments.  ... 
doi:10.18653/v1/p19-1028 dblp:conf/acl/LiS19 fatcat:fd267yehgnggphq7jtznfzi2la

Granular computing, computational intelligence, and the analysis of non-geometric input spaces

Lorenzo Livi, Alireza Sadeghian
2015 Granular Computing  
As a conclusion, we elaborate over the fundamental, conceptual problems underlying the process of data granulation, which drive the quest for a sound theory of granular computing.  ...  Data granulation emerged as an important paradigm in modeling and computing with uncertainty, exploiting information granules as the main mathematical constructs involved in the context of granular computing  ...  A particularly interesting consequence of this design choice is the fact that a granular neural network typically produces a granular output, hence consistent with the framework chosen for the IGs.  ... 
doi:10.1007/s41066-015-0003-0 fatcat:3comtfxeobda3ofim5qmuomg6i

Augmenting Neural Networks with First-order Logic [article]

Tao Li, Vivek Srikumar
2020 arXiv   pre-print
Our framework systematically compiles logical statements into computation graphs that augment a neural network without extra learnable parameters or manual redesign.  ...  In this paper, we present a novel framework for introducing declarative knowledge to neural network architectures in order to guide training and prediction.  ...  Acknowledgements We thank members of the NLP group at the University of Utah for their valuable insights and suggestions; and reviewers for pointers to related works, corrections, and helpful comments.  ... 
arXiv:1906.06298v3 fatcat:o4i7fbgvmrc5hlmi5zhoawfg7q

Joint Constrained Learning for Event-Event Relation Extraction [article]

Haoyu Wang, Muhao Chen, Hongming Zhang, Dan Roth
2021 arXiv   pre-print
Due to the lack of jointly labeled data for these relational phenomena and the restriction on the structures they articulate, we propose a joint constrained learning framework for modeling event-event  ...  Specifically, the framework enforces logical constraints within and across multiple temporal and subevent relations by converting these constraints into differentiable learning objectives.  ...  Acknowledgement We appreciate the anonymous reviewers for their insightful comments.  ... 
arXiv:2010.06727v2 fatcat:lrruts6vhvewlab5yxupltit6q

Neural-Symbolic Integration for Interactive Learning and Conceptual Grounding [article]

Benedikt Wagner, Artur d'Avila Garcez
2022 arXiv   pre-print
Interaction with the user then confirms or rejects a revision of the neural model using logic-based constraints that can be distilled into the model architecture.  ...  Neural-symbolic integration and explanation allow users and domain-experts to learn about the data-driven decision making process of large neural models.  ...  Following querying, the neural model can be constrained based on a user selection of logical formulas Knew for further learning.  ... 
arXiv:2112.11805v2 fatcat:2sgfphqpqfhgtc2kplnigcgkwa

SOFT-COMPUTING TECHNIQUES FOR FUNCTIONAL MRI BRAIN IMAGE CLASSIFICATION

Dr Mohan Awasthy*1, Dr Moon Banarjee2
2020 Zenodo  
The objective of this paper is to explore Artificial Neural Network (ANN) with Hybrid Model with reduced set of features and its comparison with various existing model.  ...  A model with high classification accuracy will be tried to develop.  ...  A neuro-fluffy framework dependent on a basic fluffy framework is prepared by methods for an information driven taking in technique got from neural system hypothesis.  ... 
doi:10.5281/zenodo.4010258 fatcat:vkykck6ksfbatfaig5yqsn5qme

A Neuro-Fuzzy Case Based Reasoning Framework for Detecting Lassa Fever Based on Observed Symptoms

Samuel Ekene Nnebe, Nora Augusta Ozemoya Okoh, Adetokunbo Mac Gregor John-Otumu, Emmanuel Osaze Oshoiribhor
2019 American Journal of Artificial Intelligence  
Based on these information gathered, the authors decided to design a hybridized intelligent framework driven by the integration of Neural Network (NN), Fuzzy logic (FL) and Case Based Reasoning (CBR) based  ...  Early diagnosis and treatment of Lassa fever is very vital for survival.  ...  for diagnoses available, control measures, degree of severity classification, possible treatment plan for suspected cases, and finally some of their published work on the trend of Lassa fever.  ... 
doi:10.11648/j.ajai.20190301.12 fatcat:6lilubofsjaxfmwgk4dgqixy7u

NEVESIM: event-driven neural simulation framework with a Python interface

Dejan Pecevski, David Kappel, Zeno Jonke
2014 Frontiers in Neuroinformatics  
Altogether, the intended purpose of the NEVESIM framework is to provide a basis for further extensions that support simulation of various neural network models incorporating different neuron and synapse  ...  NEVESIM is a software package for event-driven simulation of networks of spiking neurons with a fast simulation core in C++, and a scripting user interface in the Python programming language.  ...  models in an event-driven framework for real-time applications.  ... 
doi:10.3389/fninf.2014.00070 pmid:25177291 pmcid:PMC4132371 fatcat:l27ob3vuqracfj4tdsfeomxqvi

A Review of Life Prediction Methods for PEMFCs in Electric Vehicles

Aihua Tang, Yuanhang Yang, Quanqing Yu, Zhigang Zhang, Lin Yang
2022 Sustainability  
In this paper, a number of PEMFC life prediction methods for electric vehicles are reviewed and summarized.  ...  Based on this review, the reader can also easily understand the research status of PEMFC life prediction methods and this review lays a theoretical foundation for future research.  ...  ANFIS combines a neural network with a fuzzy system to increase the influence of logic and prior knowledge and employs an artificial neural network to learn the membership function of fuzzy logic [36]  ... 
doi:10.3390/su14169842 fatcat:nfqtgjftvrgs7a32l3axm3cpcy

Neuro-symbolic Natural Logic with Introspective Revision for Natural Language Inference

Yufei Feng, Xiaoyu Yang, Xiaodan Zhu, Michael Greenspan
2022 Transactions of the Association for Computational Linguistics  
We introduce a neuro-symbolic natural logic framework based on reinforcement learning with introspective revision.  ...  The framework is supported by properly designed local relation models to avoid input entangling, which helps ensure the interpretability of the proof paths.  ...  We thank the anonymous reviewers and action editors for their helpful comments.  ... 
doi:10.1162/tacl_a_00458 fatcat:6wfzqtiuszbddf2tjprmujaqwe

Detecting Urban Transport Modes Using a Hybrid Knowledge Driven Framework from GPS Trajectory

Rahul Das, Stephan Winter
2016 ISPRS International Journal of Geo-Information  
In this paper, a novel hybrid knowledge driven framework is developed by integrating a fuzzy logic and a neural network to complement each other's limitations.  ...  Tests demonstrate that a hybrid knowledge driven model works better than a purely knowledge driven model and at per the machine learning models in the context of transport mode detection.  ...  Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/ijgi5110207 fatcat:mdhmsldpqzftnm7g3ebpxypkky
« Previous Showing results 1 — 15 out of 41,835 results