Sign in

paulbogdan.bsky.social

@paulbogdan.bsky.social
10 followers 7 following 11 posts
PostsRepliesMedia
paulbogdan.bsky.social @paulbogdan.bsky.social · 28/09/2026
How does an intelligent agent discover which parts of its environment it can actually control? Can observations distinguish causal influence from misleading correlations? “Discovering What You Can Control: Interventional Boundary Discovery for Reinforcement Learning,” lnkd.in/ggXMCJ8f
lnkd.in
LinkedIn
This link will take you to a page that’s not on LinkedIn
000
paulbogdan.bsky.social @paulbogdan.bsky.social · 27/09/2026
Heading to @NITMB for "Data-driven modeling, simulation and inference for neurobiological problems"! 🧠📈 I'll share how we infer the hidden dynamics of living neuronal networks from scarce data, and what they reveal about NeuroAI and LLMs. Great lineup, can't wait! www.nitmb.org/event/data-d...
nitmb.org
Data-driven modeling, simulation and inference for neurobiological problems - National Institute for Theory and Mathematics in Biology (NITMB)
010
paulbogdan.bsky.social @paulbogdan.bsky.social · 22/09/2026
Biological Geometry-aware AI (BGAI) Detects Sleep Apnea Severity from Just Ten Minutes! www.nature.com/articles/s41...
nature.com
Long-range interdependencies aware signal analysis for rapid and accurate obstructive sleep apnea prediction - Nature Communications
Researchers develop a framework that captures long-range physiological interactions from sleep recordings to predict obstructive sleep apnea severity. Validated in more than 35,000 individuals, it sho...
000
paulbogdan.bsky.social @paulbogdan.bsky.social · 01/09/2026
Mastering Complexity Is the Prerequisite for Designing the Next Generation of AI and the Materials That Will Power It! "Synthesis, Self-Assembly, Graph Theory, and Applications of Complex Nanoparticle Systems" pubs.acs.org/chreay/artic...
pubs.acs.org
Synthesis, Self-Assembly, Graph Theory, and Applications of Complex Nanoparticle Systems
Abstract. The engineering of nanoparticles and their assemblies has traditionally focused on creating highly periodic systems with properties that match pr
010
paulbogdan.bsky.social @paulbogdan.bsky.social · 14/05/2026
www.science.org/doi/10.1126/... represents a step toward advancing our understanding of complex systems and developing new tools for AI and biological intelligence. Thank you very much to National Science Foundation (NSF) and University of Michigan COMPASS: Center for Complex Particle Systems!
science.org
Decoding collective dynamics and complexity in nanoparticle assemblies using graph theory
Being intermediate in scale between molecules and colloids, nanoparticles combine characteristics of both. The structure of their self-assembled states combining order and disorder is difficult to qua...
031
Reposted by @paulbogdan.bsky.social
Changdae Oh @changdaeoh.bsky.social · 07/02/2026
Have been cooking with so many amazing professionals, Seongheon Park, @teknology.bsky.social , Jiatong Li, Wendi Li, Samuel Yeh, Xuefeng Du, Hamed Hassani, @paulbogdan.bsky.social , Dawn Song, and Sharon Li🥰 Truly enjoyed it and learned a lot🙏
061
paulbogdan.bsky.social @paulbogdan.bsky.social · 21/12/2025
Graph theory and multifractal graph analysis can explore the thermodynamic parameters among elements, identify possible favorable (miscible) pairs and their congeners and enable new artificial intelligence frameworks for engineering complexity and material discovery. doi.org/10.1002/advs...
doi.org
Graph‐Theory Approach to Element Miscibility and Alloy Design
Graph and network theory enables pathway toward complex multiscale interactions between different elements for alloy design or interface engineering. Utilizing element's inherent properties and prefe....
020
paulbogdan.bsky.social @paulbogdan.bsky.social · 28/11/2025
🎉 Excited to share our NeurIPS 2025 paper on circuit representation learning! MIHC bridges topological netlist and geometric layout information for chip design tasks. An important step toward trustworthy AI in Electronic Design Automation. 📄 Read the full paper: openreview.net/pdf?id=YFQ0X...
openreview.net
010
paulbogdan.bsky.social @paulbogdan.bsky.social · 08/11/2025
Thank you COMPASS: compass.engin.umich.edu/spiking-dyna...
compass.engin.umich.edu
Exploring Brain Complexity: Professor Paul Bogdan’s Breakthrough in Nature Communications | COMPASS
000
paulbogdan.bsky.social @paulbogdan.bsky.social · 07/10/2025
Fundamental to our existence, agriculture relies on consistent seed flow for seeding and seed collection. Over-reliance on toxic solid lubricants harms our ecosystems, pollinators and all of us. authors.elsevier.com/c/1lurr_wvIm... shows graph theory enable the design of benign solid lubricant !
authors.elsevier.com
000
paulbogdan.bsky.social @paulbogdan.bsky.social · 31/07/2025
Faced with brain sensing limitations, can we + Decipher the networks from neuronal spiking behavior? + Infer neuronal networks that are resilient to variations in circuit inputs? + Distinguish networks of cognitive tasks? Congratulations to Ruochen Yang, Roozbeh Kiani, Heng Ping and Xiongye Xiao !
nature.com
Spiking dynamics of individual neurons reflect changes in the structure and function of neuronal networks - Nature Communications
Neuronal firing shows complex dynamical patterns not captured by conventional statistics. Here, authors perform the multifractal analysis of spike trains to reveal hidden temporal structures linked to...
000
paulbogdan.bsky.social @paulbogdan.bsky.social · 18/07/2025
How can we control non-Markovian dynamical systems? This ICML 2025 paper icml.cc/virtual/2025... or arxiv.org/abs/2502.04649 introduces a novel end-to-end machine learning framework for solving non-Markovian optimal control problems.
lnkd.in
LinkedIn
This link will take you to a page that’s not on LinkedIn
020