Filters








1,756 Hits in 7.3 sec

Neuro-symbolic representation learning on biological knowledge graphs

Mona Alshahrani, Mohammad Asif Khan, Omar Maddouri, Akira R Kinjo, Núria Queralt-Rosinach, Robert Hoehndorf, Janet Kelso
2017 Bioinformatics  
Results: We develop a novel method for feature learning on biological knowledge graphs.  ...  In the past years, feature learning methods that are applicable to graph-structured data are becoming available, but have not yet widely been applied and evaluated on structured biological knowledge.  ...  Acknowledgements A prototype of the feature learning algorithm was implemented at the NBDC/DBCLS BioHackathon 2016 in Tsuruoka.  ... 
doi:10.1093/bioinformatics/btx275 pmid:28449114 pmcid:PMC5860058 fatcat:6henlhfbgvee3m74izxt4rgcmq

Fast and scalable learning of neuro-symbolic representations of biomedical knowledge [article]

Asan Agibetov, Matthias Samwald
2018 arXiv   pre-print
Based on a recently published comprehensive biological knowledge graph (Alshahrani, 2017) that was used for demonstrating neuro-symbolic representation learning, we show how to train fast (under 1 minute  ...  In this work we address the problem of fast and scalable learning of neuro-symbolic representations for general biological knowledge.  ...  neuro-symbolic representations amenable for down-stream use in machine learning algorithms.  ... 
arXiv:1804.11105v1 fatcat:xnq3qva3qbg4fcn6twoyhpkele

Neuro-Symbolic Learning: Principles and Applications in Ophthalmology [article]

Muhammad Hassan, Haifei Guan, Aikaterini Melliou, Yuqi Wang, Qianhui Sun, Sen Zeng, Wen Liang, Yiwei Zhang, Ziheng Zhang, Qiuyue Hu, Yang Liu, Shunkai Shi (+15 others)
2022 arXiv   pre-print
Thus, the neuro-symbolic learning (NeSyL) notion emerged, which incorporates aspects of symbolic representation and bringing common sense into neural networks (NeSyL).  ...  Attempts have been made to overcome the challenges in neural network computing by representing and embedding domain knowledge in terms of symbolic representations.  ...  The neuro-symbolic concept learning (NS-CL) [48] built on object-based scene representation and translation into an executable symbolic program can learn visual concepts and semantic information without  ... 
arXiv:2208.00374v1 fatcat:pktmnomj3bbwpjyj7lmu37rl7i

Neuro-Symbolic Artificial Intelligence: Current Trends [article]

Md Kamruzzaman Sarker, Lu Zhou, Aaron Eberhart, Pascal Hitzler
2021 arXiv   pre-print
Neuro-Symbolic Artificial Intelligence -- the combination of symbolic methods with methods that are based on artificial neural networks -- has a long-standing history.  ...  The article is meant to serve as a convenient starting point for research on the general topic.  ...  , or whether it is primarily based on neural representations of symbolic knowledge. ( 8 ) Learning versus reasoning -this refers to the core functionality of the system, namely whether its focus is on  ... 
arXiv:2105.05330v2 fatcat:4rmmoudmtvhbbhjpvuza6r4btm

Survey on Applications of Neurosymbolic Artificial Intelligence [article]

Djallel Bouneffouf, Charu C. Aggarwal
2022 arXiv   pre-print
This success is due to its stellar performance combined with attractive properties, such as learning and reasoning.  ...  The new emerging Neurosymbolic field is currently experiencing a renaissance, as novel frameworks and algorithms motivated by various practical applications are being introduced, building on top of the  ...  The former employs fuzzy rules based on knowledge elicited from experts. The latter is based on neurules, a type of neuro-symbolic rules that combine a symbolic and a connectionist representation.  ... 
arXiv:2209.12618v1 fatcat:v37shgkahzdgznmob2pq4qimli

Is Neuro-Symbolic AI Meeting its Promise in Natural Language Processing? A Structured Review [article]

Kyle Hamilton, Aparna Nayak, Bojan Božić, Luca Longo
2022 arXiv   pre-print
Advocates for Neuro-Symbolic Artificial Intelligence (NeSy) assert that combining deep learning with symbolic reasoning will lead to stronger AI than either paradigm on its own.  ...  We find that systems where logic is compiled into the neural network lead to the most NeSy goals being satisfied, while other factors such as knowledge representation, or type of neural architecture do  ...  [15] survey the area of neuro-symbolic reasoning on Knowledge Graphs (KGs).  ... 
arXiv:2202.12205v2 fatcat:s4qkntpambct7njmaqp42y2hvu

Machine learning with biomedical ontologies [article]

Maxat Kulmanov, Fatima Zohra Smaili, Xin Gao, Robert Hoehndorf
2020 biorxiv/medrxiv   pre-print
Ontologies have long been employed in the life sciences to formally represent and reason over domain knowledge, and they are employed in almost every major biological database.  ...  Recently, ontologies are increasingly being used to provide background knowledge in similarity-based analysis and machine learning models.  ...  Neuro-symbolic systems and neuro-symbolic integration [82, 83] provide a framework in which machine learning is integrated with symbolic representations; in the neuro-symbolic cycle, deductive inference  ... 
doi:10.1101/2020.05.07.082164 fatcat:wpy4r3v7cjen7ehlqdrjssuj64

Explainable Biomedical Recommendations via Reinforcement Learning Reasoning on Knowledge Graphs [article]

Gavin Edwards, Sebastian Nilsson, Benedek Rozemberczki, Eliseo Papa
2021 arXiv   pre-print
In other domains, a neurosymbolic approach of multi-hop reasoning on knowledge graphs has been shown to produce transparent explanations.  ...  The approach is found to outperform the best baselines by 21.7% on average whilst producing novel, biologically relevant explanations.  ...  ., 2019) and pLogicNet (Qu and Tang, 2019) both involve the learning of rules. Recent updates to AnyBURL added a more neuro-symbolic style.  ... 
arXiv:2111.10625v1 fatcat:7bj5qstez5er5gqdszneidrcdq

Semantic similarity and machine learning with ontologies

Maxat Kulmanov, Fatima Zohra Smaili, Xin Gao, Robert Hoehndorf
2020 Briefings in Bioinformatics  
Ontologies have long been employed in the life sciences to formally represent and reason over domain knowledge and they are employed in almost every major biological database.  ...  Recently, ontologies are increasingly being used to provide background knowledge in similarity-based analysis and machine learning models.  ...  Neuro-symbolic systems and neuro-symbolic integration [127, 128] provide a framework in which machine learning is integrated with symbolic representations; in the neuro-symbolic cycle, deductive inference  ... 
doi:10.1093/bib/bbaa199 pmid:33049044 pmcid:PMC8293838 fatcat:3mqrjqnggrhdrkvsl6w4odazeu

Integrating Logical Rules Into Neural Multi-Hop Reasoning for Drug Repurposing [article]

Yushan Liu, Marcel Hildebrandt, Mitchell Joblin, Martin Ringsquandl, Volker Tresp
2020 arXiv   pre-print
The graph structure of biomedical data differs from those in typical knowledge graph benchmark tasks.  ...  We apply our method to the biomedical knowledge graph Hetionet and show that our approach outperforms several baseline methods.  ...  Conclusion We have proposed a novel neuro-symbolic knowledge graph reasoning approach that leverages path-based reasoning, representation learning, and logical rules.  ... 
arXiv:2007.05292v1 fatcat:sd5ftnax7jawdll3kndhx7slxy

Deep Node Ranking for Neuro-symbolic Structural Node Embedding and Classification [article]

Blaž Škrlj, Jan Kralj, Janez Konc, Marko Robnik-Šikonja, Nada Lavrač
2021 arXiv   pre-print
Furthermore, it enables exploration of the relationship between symbolic and the derived sub-symbolic node representations, offering insights into the learned node space structure.  ...  to state-of-the-art approaches on 15 real-life node classification benchmarks.  ...  Albeit being actively explored, the notion of neuro-symbolic representation learning was, to our knowledge, not yet considered in the context of node representation learning, which is the key focus of  ... 
arXiv:1902.03964v6 fatcat:zlwkh66cqrclpiydbsr2ckdcf4

From Statistical Relational to Neuro-Symbolic Artificial Intelligence

Luc de Raedt, Sebastijan Dumančić, Robin Manhaeve, Giuseppe Marra
2020 Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence  
Neuro-symbolic and statistical relational artificial intelligence both integrate frameworks for learning with logical reasoning.  ...  These cannot only be used to characterize and position neuro-symbolic artificial intelligence approaches but also to identify a number of directions for further research.  ...  Generally, knowledge graph embedding is a technique used to complete the knowledge graph triplets.  ... 
doi:10.24963/ijcai.2020/677 dblp:conf/ijcai/DongHWS020 fatcat:srd5r66dovefrpa5drxmrek25e

Directions for Explainable Knowledge-Enabled Systems [article]

Shruthi Chari, Daniel M. Gruen, Oshani Seneviratne, Deborah L. McGuinness
2020 arXiv   pre-print
As Artificial Intelligence models have become more complex, and often more opaque, with the incorporation of complex machine learning techniques, explainability has become more critical.  ...  Neuro-Symbolic AI Methods Neuro-Symbolic integration is a hybrid field that marries inductive and statistical learning capabilities of ML methods with the symbolic and conceptual representation capabilities  ...  [44] , identify tasks that will benefit from a Neuro-Symbolic Integration, including knowledge acquisition, fuzzy reasoning, and interpreting deep learning methods.  ... 
arXiv:2003.07523v1 fatcat:mnsqhmeq6nfttirizkt5mipvwy

The Third AI Summer: AAAI Robert S. Engelmore Memorial Lecture

Henry Kautz
2022 The AI Magazine  
Englemore Memorial Lecture presented at the Thirty-Fourth AAAI Conference on Artificial Intelligence on February 10, 2020.  ...  Conference on Representation Learning."  ...  Later in this essay, we will describe a different family of graph-based knowledge representation formalisms called "graphical models" that combine logic, graph theory, and probability theory.  ... 
doi:10.1609/aimag.v43i1.19122 fatcat:vrymeyxjdbhr3etnvdegqxjypa

An energy-based model for neuro-symbolic reasoning on knowledge graphs [article]

Dominik Dold, Josep Soler Garrido
2021 arXiv   pre-print
Machine learning on graph-structured data has recently become a major topic in industry and research, finding many exciting applications such as recommender systems and automated theorem proving.  ...  The presented model is mappable to a biologically-inspired neural architecture, serving as a first bridge between graph embedding methods and neuromorphic computing - uncovering a promising edge application  ...  A widely adapted approach of making the symbolic elements of graphs accessible to machine learning methods are graph embedding algorithms [4] - [6] , where nodes and edges of a graph are mapped into  ... 
arXiv:2110.01639v1 fatcat:sh7s2uiqvngz5l5niry7cutavi
« Previous Showing results 1 — 15 out of 1,756 results