Levente Szabados

Welcome to read the article "Integrated Gradients for Feature Assessment in Point Cloud-Based Data Sets" which is written by Mr. Markus Schwegler, Prof. Christoph Müller and Alexander Reiterer. This excellent article was published in issue 7 in 2023 and has been viewed 1313 times and cited 3 times! Article link: https://lnkd.in/gMCsed5z MDPI Fraunhofer IPM Furtwangen University Albert-Ludwigs-Universität Freiburg (Albert Ludwig University of Freiburg) #pointcloud #neuralnetwork #integratedgradients

2

NEW #KempnerInstitute Deeper Learning blog: How does critical batch size for pre-training scale with model and data size? We empirically and theoretically show that it scales with data size, while being almost invariant to model size. Read on: https://bit.ly/4eGMN8M

2

Implementing Anthropics Contextual Retrieval for Powerful RAG Performance This article will show you how to implement the contextual retrieval idea proposed by AnthropicContinue reading on Towards Data Science »... https://lnkd.in/e-kkhwYv #AI #ML #Automation

1

Federated Evaluation made easy with Flower 🌸 We dived into this topic with new docs and code on the 2️⃣ main approaches to evaluate models in FL systems: centralized (or server-side) evaluation ➕ federated (or client-side) evaluation 💻🔥 🧭 Explore how to implement both approaches: https://buff.ly/3UIZ4m4 👩🏽‍💻 Find code example here: https://buff.ly/3rBo7fv Do you need any clarification? Want to share your knowledge? Join the 🌼 Flower community today: https://buff.ly/3YONpTy or 👤 Log in to Flower Discuss: https://buff.ly/3ThujmD

22

🚀 Exciting News! 🚀 We are thrilled to announce that Kumo AI's position paper "Relational Deep Learning: Graph Representation Learning on Relational Databases," has been accepted at #ICML 2024! 🎉 Join us in Vienna from July 21-27, where our very own Matthias Fey will be presenting this exciting new research area. 🔗 https://icml.cc/ In collaboration with Stanford University, we introduce Relational Deep Learning (#RDL), a new research area for applying deep learning on relational databases by viewing them as temporal, heterogeneous graphs. This method eliminates the need for manual feature engineering, leading to more accurate models built faster. 🚀 We’ve also developed #RelBench, a benchmark suite to facilitate research in this new area, using diverse datasets from Stack Exchange discussions to Amazon book reviews. 📚 Stay tuned for more updates on our presentation format, and check out our paper here: https://lnkd.in/dk24vbSg #ICML2024 #MachineLearning #AI #GraphNeuralNetworks #RelationalDeepLearning #DataScience #KumoAI #StanfordCollaboration

38

Our next video is out! This is the first of a 2-part video in which Prof. Semih Salihoğlu explains about Resource Description Framework (RDF), a popular data model in graph DBMSs. If you're curious about graph modeling and are intimidated by the terminology used (especially when it comes to RDF), this video has you covered! ✅ Fundamentals of RDF: What are resources, URIs, literals, and (subject, predicate, object) triples, and how do these form a graph? ✅ Advantages of using RDF as a data model for complex and irregular domains and its flexibility in handling irregularities and connections between objects ✅ How RDF allows you to query both data and schema information in a uniform way. In the 2nd video (coming soon!), we will also go over RDF's powerful knowledge representation and reasoning capabilities, and their potential in the age of AI. Stay tuned! https://lnkd.in/gZM79U44

30

Today, we’ll explore a new development in #RAG: the addition of effective modules aimed at solving several key challenges within RAG. Retrieval-augmented generation (RAG) systems have significantly improved the performance of large language models (LLMs) by integrating external knowledge sources. However, current RAG systems still face several critical challenges: > Information Plateau: Generating a single query for information retrieval limits the scope of retrievable information, often leading to incomplete answers. > Ambiguity: Misalignment between the input question and the retrieval query due to ambiguous phrasing hinders accurate interpretation and retrieval. > Irrelevant Knowledge: Retrieved information often includes irrelevant data, which introduces noise and reduces the quality of responses. > Redundant Retrieval: Repeated retrieval of the same information for similar queries leads to inefficiency. Know more. https://lnkd.in/gnAPcZZF

2

Tomorrow! The first 2024 VRLS course is "Causal Moderated Mediation Analysis - A Causal Investigation of Heterogeneity in Mechanisms: Methods and Software." There's still time to sign up before tomorrow! https://lnkd.in/ecauvw-x

3

A group of #AIDAVA and RES-Q+ ontology experts met in Freiburg on November 11 and 12, in the Averbis GmbH offices. The objective was to assess the ontologies being currently developed in both projects and how an upper level ontology – based on SIO [https://lnkd.in/ebDPGktf] and BTL2 [https://lnkd.in/egcVA7Qf] – could help solve the different issues encountered and potentially align ontologies of both projects. The team successfully drafted a “Standard Upper-Level Ontology” – called SULO - and tested its soundness and fitness, working through several examples, demonstrating the value of such an approach. Both teams will further work on testing SULO and come with recommendations on how to further use such an approach within the respective projects but also on how it could be used for other projects. The long-term aim is to agree develop methodological approach – and a supporting Upper-Level Ontology - to develop domain specific ontologies in alignment. 

29

The Rails 8 feature set perfectly complements SQLite's power in creating resilient, high-performance production apps, but still the question lingers: Can I really go all-in on #SQLite? Join Stephen Margheim at #RailsWorld as he illustrates how to leverage Rails and SQLite's full potential in your next venture, and when SQLite does and doesn't make sense for your application. https://lnkd.in/eaBstScy

52

1 Kommentar

We are pleased to announce the publication of our inaugural blog post. In "Understanding Decision Trees", Maximilian Schneiderbauer presents a comprehensive analysis of this frequently neglected foundational model in machine learning. https://lnkd.in/eifaSREU

4

Here there's a new paper on using a decision tree model to estimate audio source locations by Paweł Antoniuk. While deep learning is popular, this study shows that traditional methods can still achieve great results - especially with binaural audio, used for immersive sound in headphones. As spatial audio becomes more common, tools like this help us better understand and analyze sound. Check it out in the EURASIP Journal on Audio, Speech, and Music Processing. #musictech #audiotech #machinelearning #ml

20

🆕 New Blog By Shadia Nassar,PhD Dive into the "contribution analysis" world with Includovate's latest blog! ✍ This blog explores the fascinating concept of determining how different things contribute to a specific outcome. Unlike causal analysis (➡️), contribution analysis assigns weights to various influencing factors to understand their impact. 🤔 But how do you build a convincing claim about contribution? What role does causality play in contribution analysis ❓ how does it help stakeholders make informed decisions❓ 👇 Explore practical examples of contribution analysis in evaluation, from analysing public health interventions to evaluating health sciences programmes ‍. 👇Discover how this structured approach helps reduce uncertainty and provides valuable insights into programme effectiveness. 👇Uncover the ethical complications of contribution analysis in the AI revolution and the impact of algorithms on societal inequalities ⚖️. 👇Join the discussion ️ and learn how to navigate the complexities of contribution analysis to achieve greater success in your goals ! https://lnkd.in/dsaqq8WV #contributionanalysis #evaluation #Includovate #blogs #decisionmaking

14

4 Kommentare

In this week's D-Lab Blog Post, Senior Data Scientist and Curriculum Lead Tom van Nuenen makes the case for why we need Digital Hermeneutics. What is Hermeneutics? In Tom's words: "Hermeneutics is just a fancy word for interpretation, both in the practical sense of a sustained methodological effort, and in the philosophical sense of asking what interpretation is in the first place." This is particularly important given the rise of Large Language Models - which Tom discusses in his piece! 👇 Check it out on Medium, and subscribe for weekly blog posts on data science! https://lnkd.in/g4RDEQaM

2