Sergei RYBALKO

I've made it to the top 36-54 researchers globally in terms of the number of accepted papers to NeurIPS 2024. Our papers: 1. Poster On the Optimal Time Complexities in Decentralized Stochastic Asynchronous Optimization Alexander Tyurin · Peter Richtarik 2. Poster Shadowheart SGD: Distributed Asynchronous SGD with Optimal Time Complexity Under Arbitrary Computation and Communication Heterogeneity Alexander Tyurin · Marta Pozzi · Ivan Ilin · Peter Richtarik 3. Poster MicroAdam: Accurate Adaptive Optimization with Low Space Overhead and Provable Convergence Ionut-Vlad Modoranu · Mher Safaryan · Grigory Malinovsky · Eldar Kurtić · Thomas Robert · Peter Richtarik · Dan Alistarh 4. Poster Freya PAGE: First Optimal Time Complexity for Large-Scale Nonconvex Finite-Sum Optimization with Heterogeneous Asynchronous Computations Alexander Tyurin · Kaja Gruntkowska · Peter Richtarik 5. Oral PV-Tuning: Beyond Straight-Through Estimation for Extreme LLM Compression Vladimir Malinovskii · Denis Mazur · Ivan Ilin · Denis Kuznedelev · Konstantin Burlachenko · Kai Yi · Dan Alistarh · Peter Richtarik 6. Poster Don't Compress Gradients in Random Reshuffling: Compress Gradient Differences Abdurakhmon Sadiev · Grigory Malinovsky · Eduard Gorbunov · Igor Sokolov · Ahmed Khaled · Konstantin Burlachenko · Peter Richtarik 7. Poster The Power of Extrapolation in Federated Learning Hanmin Li · Kirill Acharya · Peter Richtarik 8. Spotlight Improving the Worst-Case Bidirectional Communication Complexity for Nonconvex Distributed Optimization under Function Similarity Kaja Gruntkowska · Alexander Tyurin · Peter Richtarik 9. Poster Byzantine Robustness and Partial Participation Can Be Achieved at Once: Just Clip Gradient Differences Grigory Malinovsky · Peter Richtarik · Samuel Horváth · Eduard Gorbunov --- Disclaimer: It is quality that matters, not quantity. I am very proud of the stuff we achieved in our papers. More papers typically reflect the work of a larger group. Nevertheless, an interesting plot to look at. --- Source: https://lnkd.in/egURv5s8 NeuriPS is the main conference in the field of AI (https://lnkd.in/eTeEJgdD).

282

16 commentaires

We all know that researchers spend a significant amount of time searching arXiv for relevant papers. In our spare time, we decided to tackle this problem head-on. Introducing AI-powered arXiv Search 🔥 🔍 : A tool that can quickly pinpoint and summarize research topics from arXiv, cutting down the time spent sifting through countless papers. After spending 14 hours vectorizing abstracts from arXiv papers up to 01/12/2024, we integrated them into a Retrieval-Augmented Generation (RAG) system. We now leverage PaperQA as the RAG engine, improving upon the traditional in-memory approach by utilizing a dedicated vector database of abstracts. It’s fully open-source — including the vector DB. Check it out and let us know what you think! https://lnkd.in/eme_mjCN #RAG #ArXiv #PaperQA

16

Souriau's Geometric Principles for Quantum Mechanics "It's quantum so it's geometric" - Jean-Marie Souriau The symplectic geometry of statistical mechanics was constructed by Jean-Marie Souriau to address quantum mechanics. A geometric reinterpretation of the states of statistical mechanics makes it possible to introduce the axiomatics of quantum states and solutions of a simple set of inequalities. This axiomatic provides a probabilistic interpretation, relating the Hilbert space and the unitary representation of a group called quantum representation. GDR CNRS IASIS: Structure and geometric definition of Entropy as a Casimir function of a symplectic foliation: extension of informational measurement to Lie groups and quantum mechanics by Thales at Institut Henri Poincaré Full slides: https://lnkd.in/ebhJuiaf GDR CNRS IASIS DAY: https://lnkd.in/eDr-FnZb

4

Advances in Bayesian Deep Neural Network Ensembles and Active Learning for Preference Modeling Quick read: https://lnkd.in/gzZB2VGh Paper 1: https://lnkd.in/gBkwknBZ Paper 2: https://lnkd.in/gNrqnqg9

18

👉🏼 Chapitre 7. Applications et éclairage éthique des Intelligences Artificielles Génératives linguistiques en Médecine 🤓 Cécile Monteil 👇🏻 https://lnkd.in/e4Fnfwqw 🔍 Focus on data insights: - The integration of Generative Artificial Intelligence systems like Large Language Models (LLM) is reshaping medical practices. - Physicians are required to adapt and acquire new competencies in technology for effective utilization of Gen AI. 💡 Main outcomes and implications: - The evolving ethical framework surrounding Gen AI in medicine necessitates increased ethical vigilance. - Healthcare professionals need to balance technological advancements with preserving their medical expertise and critical thinking skills. 📚 Field significance: - The digitalization of medicine emphasizes the importance of training physicians in new technologies to maintain quality healthcare services. - Gen AI presents both risks and opportunities that require careful consideration within the healthcare sector. 🗄️: [#GenerativeAI #Healthcare #EthicalFramework #DataInsights]

3

1 commentaire

[Article] Our paper "Applying non-negative tensor factorization to centered data" is finally available on BMI, Vol.174, n°3. Congratulations to the authors Paul Fogel, Christophe Geissler, Hans-Jörg von Mettenheim and George Luta! >> More information on our website: https://lnkd.in/e6WQ8iu8 #AI #ESGData #Mazars

21

⚕🤖 MEDITRON-70B (based on Llama 2) already outperforms GPT-3.5 and Med-PaLM and is within 5% of GPT-4 and 10% of Med-PaLM-2. And now, a new version that is fine-tuned from AI at Meta's Llama 3🦙🦙🦙 8B achieves promising results.

7

Some insightful and ground-breaking work about HDC from our colleague Robert McMenemy 👾! Thank you!!! "🦾Integrating Neuromorphic Processing and Hyperdimensional Computing in Semantic Networks: Unlocking New Frontiers in Cognitive Computing I'm excited to share my latest article that dives into the powerful synergy between neuromorphic processing, hyperdimensional computing [HDC], and semantic networks. This interdisciplinary approach has the potential to revolutionize how we model cognitive processes, handle complex data, and build adaptive AI systems. In the article, I explore how combining these computational paradigms can enhance cognitive modelling, knowledge representation, and reasoning capabilities. One of the standout applications is in ontology development, where this integrated method can manage complex and dynamic ..." https://lnkd.in/gG9SNZwk #CognitiveComputing #NeuromorphicProcessing #HyperdimensionalComputing #SemanticNetworks #AI #MachineLearning #DataScience #KnowledgeRepresentation #OntologyDevelopment #AdaptiveAI #BigData #ComplexData #AIApplications #InterdisciplinaryResearch #CognitiveModelling #Robotics #ArtificialIntelligence #PatternRecognition #DataMining #EnergyEfficiency #BiologicallyInspiredAI #Scalability #Efficiency #NoiseTolerance #LearningAlgorithms #SubSymbolicAI #SymbolicAI #MemoryManagement #Clustering #AIResearch #ZscaleLabs

9

As Yann Lecun says, intelligence is not correlated with a desire to dominate. It follows that the more intelligent AI becomes, the less it is likely to want to dominate us. Thus, we have nothing to fear. However...that doesn't mean that a desire to dominate could not drive some future AI. It just means that any such desire would not emerge from its intelligence. Indeed, a 'dominating AI' wouldn't have to be great at math and logic, just at manipulating others. Perhaps such an AI could also, ironically, easily manipulate the most intelligent AI's - and humans - precisely because the latter don't care about domination. They'll be too busy solving quantum consciousness or some such intellectual challenge while the dumber, Machiavellian AI's take over the world. I'll leave it to you to ponder what kind of humans might be all too happy to cooperate with those kinds of entities. To consider whether intelligence has been an effective antidote to the manipulation and domination of the masses that led (and continues to lead) to mass suffering and death. How strong is the assumption that intelligence will be the only objective for AI development, forever? Which stakeholders control the objectives and what are their objectives? Then again, perhaps you're too smart to bother with such matters 😉 https://lnkd.in/gUbvZXmQ

2

Read our latest contribution on explainable AI here : HPExplorer: XAI Method to Explore the Relationship Between Hyperparameters and Model Performance

18

📢 Original article Sharing-Vol. 23 No. 7 💕 Title: Exploring Implicit Biological Heterogeneity in ASD Diagnosis Using a Multi-Head Attention Graph Neural Network 🤵 Authors: Hyung-Jun Moon, Sung-Bae Cho,* 🔔 Full Text: https://lnkd.in/gtMHteqU 😎Welcome to your reading! #Parkinsons #ParkinsonsDisease #Neurology #BrainHealth #MovementDisorders #Dopamine #NeurodegenerativeDiseases #Healthcare #Research #Awareness #Treatment #SupportGroups #PatientCare #Therapeutics #MentalHealth #Neuroscience #CureParkinsons #BrainMapping #Neurobiology #BrainInjury #NeuroscienceNews #NeuroImaging

3

𝐌𝐮𝐬𝐭-𝐫𝐞𝐚𝐝, 𝐟𝐫𝐞𝐞 𝐛𝐨𝐨𝐤 𝐭𝐡𝐚𝐭 𝐩𝐫𝐨𝐯𝐢𝐝𝐞𝐬 𝐚 𝐩𝐚𝐭𝐡𝐰𝐚𝐲 𝐟𝐫𝐨𝐦 𝐆𝐀𝐈 𝐭𝐨 𝐀𝐆𝐈! ​ Yann LeCun has expressed skepticism about the potential of LLMs to achieve Artificial General Intelligence (AGI), citing limitations in memory, planning, and grounding in real-world understanding. However, we argue that 𝗟𝗟𝗠 𝗖𝗼𝗹𝗹𝗮𝗯𝗼𝗿𝗮𝘁𝗶𝘃𝗲 𝗜𝗻𝘁𝗲𝗹𝗹𝗶𝗴𝗲𝗻𝗰𝗲 (𝗟𝗖𝗜)—particularly with multiple multimodal LLMs working together—offers a promising architecture to address these challenges and progress toward AGI. ​ Two key architectural innovations underlie LCI: ​ 𝐒𝐞𝐩𝐚𝐫𝐚𝐭𝐢𝐨𝐧 𝐨𝐟 𝐔𝐧𝐜𝐨𝐧𝐬𝐜𝐢𝐨𝐮𝐬 𝐏𝐫𝐨𝐜𝐞𝐬𝐬𝐢𝐧𝐠 𝐚𝐧𝐝 𝐂𝐨𝐧𝐬𝐜𝐢𝐨𝐮𝐬 𝐑𝐞𝐚𝐬𝐨𝐧𝐢𝐧𝐠 𝐚𝐧𝐝 𝐏𝐥𝐚𝐧𝐧𝐢𝐧𝐠: Inspired by the human mind’s dual-layer structure, LCI divides its architecture into foundational and adaptive layers. The foundation layer, akin to unconscious processing, leverages extensive datasets to build robust pattern recognition, encoding essential responses much like human instincts. The adaptive layer, mirroring conscious thought, enables rapid adaptation and contextual reasoning. This dual-layer approach allows LLMs to perform complex tasks with foundational knowledge while adjusting to new information on the fly, much as a child learns new concepts within a developed cognitive framework. ​ 𝐃𝐢𝐬𝐭𝐢𝐧𝐜𝐭 𝐑𝐨𝐥𝐞𝐬 𝐟𝐨𝐫 𝐄𝐚𝐜𝐡 𝐋𝐋𝐌-𝐀𝐠𝐞𝐧𝐭: Each LLM-agent is assigned a specific role, such as executive (knowledge processing), legislative (behavior guardrails), and judicial (adapting to cultural norms). ​ Additionally, our validation of the LCI framework has been thoroughly tested through real-world deployment in applications across various domains, including healthcare, sales planning, investment, and debiasing news. These empirical deployments highlight the practical viability and adaptability of the LCI framework, showcasing its ability to enhance reasoning and decision-making through collaborative intelligence. ​ The eleven aphorisms presented in this free book illustrate the philosophical foundation of the LCI framework, backed by empirical studies. Through LCI, we envision a clear pathway to AGI, rooted in collaborative intelligence and flexible, context-aware adaptation. https://lnkd.in/gn74DkNj (free) https://lnkd.in/gszSvfqP ($6.98 printing fee)

9

Journal Impact Factor JCR-WoS: 2.0 (2023) Q1 CiteScore 2023: 3.1 Q1

10

x=a[cos(x+t) +sin(x+t)]+a[sin(y+t)], y=a[sin(x+t)], z=a[cos(x+t)], r(t)=R(ωt)r_0 Complex surface topology via parametric equations, trigonometry, and rotational geometry in 3D space. This is a way of describing motion with vector calculus and kinematics. I do my best to show that math can be visually intuitive and aesthetically pleasing. #math #mathematics #mathematician #maths #physics #calculus #vectorcalculus #motion #kinematics #parametricequations #science #3danimation #blender #geometrynodes #topology #trigonometry #geometry

341

18 commentaires

Revolutionizing Quantum Tech: Scientists Achieve Collective Motion in Macroscopic Oscillators BY ECOLE POLYTECHNIQUE FÉDÉRALE DE LAUSANNE JANUARY 1, 2025 A Hexamer of Macroscopic Mechanical Oscillators Scientists at EPFL achieved a breakthrough by synchronizing six mechanical oscillators into a collective quantum state, enabling observations of unique phenomena like quantum sideband asymmetry. This advance paves the way for innovations in quantum computing and sensing.

1