GraphNews

Leveraging Structured Knowledge to Automatically Detect Hallucination in Large Language Models 🔺 🔻 Large Language Models has sparked a revolution in AI’s… | 25 comments on LinkedIn

Sometimes the words “taxonomy” and “ontology” are used interchangeably, and while they are closely related, they are not the same thing. They are both considered kinds of knowledge organization systems to support information and knowledge management.

Semantics friends and colleagues, Announcing the release of a new front-end application development library for RDF. Most of us have heard time and again how…

(This post & prompt technique was inspired by a conversation with Nicky Clarke and a recent post by Kurt Cagle, Rethinking Hypergraphs.) Building prompts for ultra-large context windows requires a complete reimagining of how prompts are created.

OccO: Occupation Ontology. Contribute to Occupation-Ontology/OccO development by creating an account on GitHub.

Structural analysis in network science is finding the information hidden from the topology structure of complex networks. Many methods have already been proposed in the research on the structural analysis of complex networks to find the different structural information of networks. In this work, the sum of nodes' betweenness centrality (SBC) is used as a new structural index to check how the structure of the complex networks changes in the process of the network's growth. We build two four different processes of network growth to check how the structure change will be manifested by the SBC. We find that when the networks are under Barabási-Albert rule, the value of SBC for each network grows like a logarithmic function. However, when the rule that guides the network's growth is the Erdős-Rényi rule, the value of SBC will converge to a fixed value. It means the rules that guide the network's growth can be illustrated by the change of the SBC in the process of the network's growth. In other words, in the structure analysis of complex networks, the sum of nodes' betweenness centrality can be used as an index to check what kinds of rules guide the network's growth.

The development of artificial intelligence systems with advanced reasoning capabilities represents a persistent and long-standing research question. Traditionally, the primary strategy to address this challenge involved the adoption of symbolic approaches, where knowledge was explicitly represented by means of symbols and explicitly programmed rules. However, with the advent of machine learning, there has been a paradigm shift towards systems that can autonomously learn from data, requiring minimal human guidance. In light of this shift, in latest years, there has been increasing interest and efforts at endowing neural networks with the ability to reason, bridging the gap between data-driven learning and logical reasoning. Within this context, Neural Algorithmic Reasoning (NAR) stands out as a promising research field, aiming to integrate the structured and rule-based reasoning of algorithms with the adaptive learning capabilities of neural networks, typically by tasking neural models to mimic classical algorithms. In this dissertation, we provide theoretical and practical contributions to this area of research. We explore the connections between neural networks and tropical algebra, deriving powerful architectures that are aligned with algorithm execution. Furthermore, we discuss and show the ability of such neural reasoners to learn and manipulate complex algorithmic and combinatorial optimization concepts, such as the principle of strong duality. Finally, in our empirical efforts, we validate the real-world utility of NAR networks across different practical scenarios. This includes tasks as diverse as planning problems, large-scale edge classification tasks and the learning of polynomial-time approximate algorithms for NP-hard combinatorial problems. Through this exploration, we aim to showcase the potential integrating algorithmic reasoning in machine learning models.

Copyright 2024. Kurt Cagle When you look at your company, you likely see things - people, clients or customers, products, processes, roles, revenue, etc.

In this paper, we present LiGNN, a deployed large-scale Graph Neural Networks (GNNs) Framework. We share our insight on developing and deployment of GNNs at large scale at LinkedIn. We present a set of algorithmic improvements to the quality of GNN representation learning including temporal graph architectures with long term losses, effective cold start solutions via graph densification, ID embeddings and multi-hop neighbor sampling. We explain how we built and sped up by 7x our large-scale training on LinkedIn graphs with adaptive sampling of neighbors, grouping and slicing of training data batches, specialized shared-memory queue and local gradient optimization. We summarize our deployment lessons and learnings gathered from A/B test experiments. The techniques presented in this work have contributed to an approximate relative improvements of 1% of Job application hearing back rate, 2% Ads CTR lift, 0.5% of Feed engaged daily active users, 0.2% session lift and 0.1% weekly active user lift from people recommendation. We believe that this work can provide practical solutions and insights for engineers who are interested in applying Graph neural networks at large scale.

The spread of misinformation poses a threat to the social media ecosystem. Effective countermeasures to mitigate this threat require that social media platforms be able to accurately detect low-credibility accounts even before the content they share can be classified as misinformation. Here we present methods to infer account credibility from information diffusion patterns, in particular leveraging two networks: the reshare network, capturing an account’s trust in other accounts, and the bipartite account-source network, capturing an account’s trust in media sources. We extend network centrality measures and graph embedding techniques, systematically comparing these algorithms on data from diverse contexts and social media platforms. We demonstrate that both kinds of trust networks provide useful signals for estimating account credibility. Some of the proposed methods yield high accuracy, providing promising solutions to promote the dissemination of reliable information in online communities. Two kinds of homophily emerge from our results: accounts tend to have similar credibility if they reshare each other’s content or share content from similar sources. Our methodology invites further investigation into the relationship between accounts and news sources to better characterize misinformation spreaders.

Connecting the dots for a theoretical analysis of Graph Neural Network models

Discover how Graph of Thoughts aims to revolutionize prompt engineering, and LLMs more broadly, enabling more flexible and human-like problem-solving.

From 5th to 9th of February, I had the privilege to participate to a seminar entitled "Are Knowledge Graphs Ready for the Real World? Challenges and Perspective" at the Dagstuhl Schlosse. It was my third Dagstuhl experience, so I was ready for a very intense week of work and discussion and I have to

KGLens: Unlocking Hidden Connections: Knowledge Graphs Reveal What Models Don't Know …. Evaluating AI Knowledge Just Got Smarter … How do we know if AI…

Graph Foundation Model (GFM) is a new trending research topic in the graph domain, aiming to develop a graph model capable of generalizing across different graphs and tasks. However, a versatile GFM has not yet been achieved. The key challenge in building GFM is how to enable positive transfer across graphs with diverse structural patterns. Inspired by the existing foundation models in the CV and NLP domains, we propose a novel perspective for the GFM development by advocating for a "graph vocabulary", in which the basic transferable units underlying graphs encode the invariance on graphs. We ground the graph vocabulary construction from essential aspects including network analysis, theoretical foundations, and stability. Such a vocabulary perspective can potentially advance the future GFM design following the neural scaling laws.

Lately I’ve become intrigued about the published research + open source code for a relatively specific topic: generating graphs to use for…

Exciting news for the data science community! PyG v2.5, the most widely used Graph Machine Learning Library, has just been released in collaboration with Intel…

Concepts form the very core of all our work on representing, accumulating, and manipulating knowledge with algorithms. Without them, we would simply have no framework to even start thinking about knowledge.

-- Source link: http://tinyurl.com/y8b8az2h-- Note: Reuters has not verified this story and does not vouch for its accuracy

Symbolic Artificial Intelligence, Semantic Web State of the art The Semantic Web project was formulated at the end of the 20th century and is based on the availability of inference engines and onto…

Entity-based SEO is the future. Discover the art of crafting search-relevant content through expert entity extraction and linking techniques.

Yes, Structured Data Markup = E-E-A-T fairy dust, or so the popular answer goes. Complete nonsense, because magic does not compute! The real story is...

The NLQ integration for Ontotext GraphDB in LangChain: a framework designed to simplify the creation of applications using LLMs

Although I have of course heard this question more often in recent months than in all the years before, it is really just a reiteration of the question of all questions, which is probably the most fundamental question of all for AI: How much human (or symbolic AI) does statistical AI need? With ever

Distributed ledger technologies have opened up a wealth of fine-grained transaction data from cryptocurrencies like Bitcoin and Ethereum. This allows research into problems like anomaly detection, anti-money laundering, pattern mining and activity clustering (where data from traditional currencies is rarely available). The formalism of temporal networks offers a natural way of representing this data and offers access to a wealth of metrics and models. However, the large scale of the data presents a challenge using standard graph analysis techniques. We use temporal motifs to analyse two Bitcoin datasets and one NFT dataset, using sequences of three transactions and up to three users. We show that the commonly used technique of simply counting temporal motifs over all users and all time can give misleading conclusions. Here we also study the motifs contributed by each user and discover that the motif distribution is heavy-tailed and that the key players have diverse motif signatures. We study the motifs that occur in different time periods and find events and anomalous activity that cannot be seen just by a count on the whole dataset. Studying motif completion time reveals dynamics driven by human behaviour as well as algorithmic behaviour.

Topological deep learning (TDL) is a rapidly evolving field that uses topological features to understand and design deep learning models. This paper posits that TDL may complement graph representation learning and geometric deep learning by incorporating topological concepts, and can thus provide a natural choice for various machine learning settings. To this end, this paper discusses open problems in TDL, ranging from practical benefits to theoretical foundations. For each problem, it outlines potential solutions and future research opportunities. At the same time, this paper serves as an invitation to the scientific community to actively participate in TDL research to unlock the potential of this emerging field.

Our new preprint, "Towards Learning on Graphs with State Space Models" is out!We present Graph Mamba Networks (GMNs), a scalable (linear time) class of Graph Neural Networks that aggregates the hierarchical neighborhoods of nodes using selective state space models.Arxiv:… pic.twitter.com/OjCHMYTvW5— Ali Behrouz (@behrouz_ali) February 14, 2024

Microsoft is transforming retrieval-augmented generation with GraphRAG, using LLM-generated knowledge graphs to significantly improve Q&A when analyzing complex information and consistently outperforming baseline RAG. Get the details.