191,971 Hits in 1.9 sec

Neural Logic Networks [article]

Shaoyun Shi, Hanxiong Chen, Min Zhang, Yongfeng Zhang
2019 arXiv   pre-print
In this paper, we propose Neural Logic Network (NLN), which is a dynamic neural architecture that builds the computational graph according to input logical expressions.  ...  It learns basic logical operations as neural modules, and conducts propositional logical reasoning through the network for inference.  ...  To integrate the advantages of deep neural networks and logical reasoning, we propose Neural Logic Network (NLN), a neural architecture to conduct logical inference based on neural networks.  ... 
arXiv:1910.08629v1 fatcat:634xfrpabrhshcl5k2irjo5tq4

Logical Neural Networks [article]

Ryan Riegel, Alexander Gray, Francois Luus, Naweed Khan, Ndivhuwo Makondo, Ismail Yunus Akhalwaya, Haifeng Qian, Ronald Fagin, Francisco Barahona, Udit Sharma, Shajith Ikbal, Hima Karanam (+3 others)
2020 arXiv   pre-print
We propose a novel framework seamlessly providing key properties of both neural nets (learning) and symbolic logic (knowledge and reasoning).  ...  Inference is omnidirectional rather than focused on predefined target variables, and corresponds to logical reasoning, including classical first-order logic theorem proving as a special case.  ...  We use the logical formulae, consisting of 5 plausible axioms, of the Smokers and Friends experiment in LTN experiment K exp2 [19] .  ... 
arXiv:2006.13155v1 fatcat:wqrrviovzvhgxmobthhej5kbve

Neural Markov Logic Networks [article]

Giuseppe Marra, Ondřej Kuželka
2020 arXiv   pre-print
We introduce neural Markov logic networks (NMLNs), a statistical relational learning system that borrows ideas from Markov logic.  ...  Like Markov logic networks (MLNs), NMLNs are an exponential-family model for modelling distributions over possible worlds, but unlike MLNs, they do not rely on explicitly specified first-order logic rules  ...  In this paper, we propose neural Markov logic networks (NMLN) .  ... 
arXiv:1905.13462v3 fatcat:vxixxxzo7nekfp3xjloatvwv3y

Computability of Logical Neural Networks

T.B. Ludermir
1992 Journal of Intelligent Systems  
Key words: computability, finite state machine, logical neural network, probabilistic automaton, weighted regular language. have the same computability power it is only necessary to prove the following  ...  We studied the computability of networks of PLNs (Probabilistic Logic Node (Aleksander, 1988)). We suggested a new method of recognition based on stored probabilities with PLN networks.  ...  Additionally, having related logical neural networks to automata, some comments can be made. Firstly, logical neural networks can be trained, from a set of examples to solve a problem.  ... 
doi:10.1515/jisys.1992.2.1-4.261 fatcat:kod2o5tqi5as5dmd3i6ymixs6a

Logic Mining Using Neural Networks [article]

Saratha Sathasivam, Wan Ahmad Tajuddin Wan Abdullah (Univ Malaya)
2008 arXiv   pre-print
Wan Abdullah [1] proposed a method of doing logic programming on a Hopfield neural network.  ...  Optimization of logical inconsistency is carried out by the network after the connection strengths are defined from the logic program; the network relaxes to neural states corresponding to a valid interpretation  ...  We extended the work related to logic programming in neural network by introducing reverse analysis method. This method is capable to induce logical rules entrenched in a database.  ... 
arXiv:0804.4071v1 fatcat:o7jbsapqezckzkyic5fiyhmv6u

Harnessing Deep Neural Networks with Logic Rules [article]

Zhiting Hu, Xuezhe Ma, Zhengzhong Liu, Eduard Hovy, Eric Xing
2020 arXiv   pre-print
Combining deep neural networks with structured logic rules is desirable to harness flexibility and reduce uninterpretability of the neural models.  ...  We propose a general framework capable of enhancing various types of neural networks (e.g., CNNs and RNNs) with declarative first-order logic rules.  ...  Related Work Combination of logic rules and neural networks has been considered in different contexts.  ... 
arXiv:1603.06318v6 fatcat:622upadxczcbfk2ulo5cncr7hm

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.  ...  Today, the dominant paradigm for training neural networks involves minimizing task loss on a large dataset.  ...  Related Work and Discussion Artificial Neural Networks and Logic Our work is related to neural-symbolic learning (e.g.  ... 
arXiv:1906.06298v3 fatcat:o4i7fbgvmrc5hlmi5zhoawfg7q

Robust Deep Neural Networks Inspired by Fuzzy Logic [article]

Minh Le
2020 arXiv   pre-print
Deep neural networks have achieved impressive performance and become the de-facto standard in many tasks.  ...  As a remedy, I propose new architectures inspired by fuzzy logic that combine several alternative design elements.  ...  To design a local and disconnected deep neural network, I take inspiration from fuzzy logic.  ... 
arXiv:1911.08635v3 fatcat:qjkpmpdfbnaqzbwmz3s76pxu7i

The Logic of Graph Neural Networks [article]

Martin Grohe
2022 arXiv   pre-print
Graph neural networks (GNNs) are deep learning architectures for machine learning problems on graphs.  ...  It has recently been shown that the expressiveness of GNNs can be characterised precisely by the combinatorial Weisfeiler-Leman algorithms and by finite variable counting logics.  ...  neural networks.  ... 
arXiv:2104.14624v2 fatcat:pkq6yfm53bfwxjznoxeegchzwi

Can recursive neural tensor networks learn logical reasoning? [article]

Samuel R. Bowman
2014 arXiv   pre-print
Recursive neural network models and their accompanying vector representations for words have seen success in an array of increasingly semantically sophisticated tasks, but almost nothing is known about  ...  To evaluate this, I train a recursive model on a new corpus of constructed examples of logical reasoning in short sentences, like the inference of "some animal walks" from "some dog walks" or "some cat  ...  [11] show in an aside that a Boolean logic with negation and conjunction can be learned in a minimal recursive neural network model with one-dimensional (scalar) representations for words.  ... 
arXiv:1312.6192v4 fatcat:b2o5hdc43vfj5gvktgnvpb5uxe

Recursive Neural Networks Can Learn Logical Semantics [article]

Samuel R. Bowman, Christopher Potts, Christopher D. Manning
2015 arXiv   pre-print
We pursue this question by evaluating whether two such models---plain TreeRNNs and tree-structured neural tensor networks (TreeRNTNs)---can correctly learn to identify logical relationships such as entailment  ...  Tree-structured recursive neural networks (TreeRNNs) for sentence meaning have been successful for many applications, but it remains an open question whether the fixed-length representations that they  ...  We conclude that TreeRNTN models are adequate for typical cases 2 Tree-structured neural networks We limit the scope of our experiments in this paper to neural network models that adhere to the linguistic  ... 
arXiv:1406.1827v4 fatcat:hye3rd434jcipcbldnyoj5nq24

Learning Algorithms via Neural Logic Networks [article]

Ali Payani, Faramarz Fekri
2019 arXiv   pre-print
We propose a novel learning paradigm for Deep Neural Networks (DNN) by using Boolean logic algebra.  ...  Logic Networks (NLNs).  ...  Introduction Deep Neural Networks (DNNs) based on Convolution Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) have improved the state of the art in various areas such as natural language processing  ... 
arXiv:1904.01554v1 fatcat:bkzn2i3uybelvpd4nvxkrwvyoi

Neural Networks Enhancement with Logical Knowledge [article]

Alessandro Daniele, Luciano Serafini
2021 arXiv   pre-print
In a previous work, we proposed KENN (Knowledge Enhanced Neural Networks), a Neural-Symbolic architecture that injects prior logical knowledge into a neural network by adding a new final layer which modifies  ...  with logic.  ...  of Neural Networks with the expressivity of First Order Logic.  ... 
arXiv:2009.06087v2 fatcat:lk4k2rvvvnfrvh7iulguh5o46u

Teaching Temporal Logics to Neural Networks [article]

Christopher Hahn, Frederik Schmitt, Jens U. Kreber, Markus N. Rabe, Bernd Finkbeiner
2021 arXiv   pre-print
We study two fundamental questions in neuro-symbolic computing: can deep learning tackle challenging problems in logics end-to-end, and can neural networks learn the semantics of logics.  ...  In this work we focus on linear-time temporal logic (LTL), as it is widely used in verification.  ...  Other works focused on recurrent neural networks or graph neural networks for code analysis, e.g.  ... 
arXiv:2003.04218v3 fatcat:bdumqav2yff6vo33yx7zha2usy

Probabilistic Logic Neural Networks for Reasoning [article]

Meng Qu, Jian Tang
2019 arXiv   pre-print
In this paper, we propose the probabilistic Logic Neural Network (pLogicNet), which combines the advantages of both methods.  ...  A principled logic rule-based approach is the Markov Logic Network (MLN), which is able to leverage domain knowledge with first-order logic and meanwhile handle their uncertainty.  ...  We propose such an approach called the probabilistic Logic Neural Networks (pLogicNet).  ... 
arXiv:1906.08495v2 fatcat:qdrrlg7jkndirhqbiweo32nbqa
« Previous Showing results 1 — 15 out of 191,971 results