English

Posts with #RepresentationLearning

#RepresentationLearning

Public posts on X that use #RepresentationLearning.

Related hashtags

  1. Elham Azizi@elhamazizi
    Check out our latest work: SHERLOCK🕵️ combines #RepresentationLearning of single-cell #perturbations with #CausalInference of their downstream transcriptional effects, supported by theoretical identifiability results 👇
    Lingting Shi @shi_lingting
    🧬 What can we really learn from a perturbation data? Can #AI uncover how perturbations are related, what they affect downstream, how effects change with context—and what about combinatorial? Introducing SHERLOCK🕵️ led by @mingZedh, Josh Myers&I @joselmcfaline, @elhamazizi 🧵👇
    22612010K
  2. Lingting Shi@shi_lingting
    🧬 What can we really learn from a perturbation data? Can #AI uncover how perturbations are related, what they affect downstream, how effects change with context—and what about combinatorial? Introducing SHERLOCK🕵️ led by @mingZedh, Josh Myers&I @joselmcfaline, @elhamazizi 🧵👇
    Download
    1219114.4K
  5. Elham Azizi@elhamazizi
    @elhamazizi
    LYNX learns a shared latent space that *links* highly heterogeneous spatial modalities (transcriptomics, proteomics, metabolomics, histology, etc) even when they differ dramatically in resolution, throughput, and feature space #RepresentationLearning
    Download
    103493
  6. Mehdijoodaki@Mehdijoodakiii
    1/5) We're extending the PILOT framework! Our new PILOT-GM-VAE is now published in Briefings in Bioinformatics. It combines Gaussian Mixtures + VAEs for more powerful patient-level analysis of single-cell atlases. #SingleCellAtlas #VAE #OptimalTransport
    Download
    236567
  7. Mehdijoodaki@Mehdijoodakiii
    @Mehdijoodakiii
    2/5) What’s new in PILOT-GM-VAE: • Gaussian Mixture VAE captures complex cell components • Efficient and high-performance Optimal Transport framework for patient-level alignment • Fully unsupervised, no cell-type labels required #RepresentationLearning #UnsupervisedLearning
    10262
  9. Meir Yossef Levi@MeirYossef_Levi
    🚀 3 papers accepted to #CVPR2026 — all on the latent space of foundation models! Universal Normal Embedding · Training-free AI video detection · Semantic invariants in CLIP. Geometry→Control→Detection→Invariance DM for details, See you in Denver #RepresentationLearning
    Download
    105360
  10. Manolis Kellis✓@manoliskellis
    The first six lectures of #DeepLearning for #Genomics and #LifeSciences are posted on @YouTube here: covering #MachineLearning #CNNs #RNNs #GNNs #RepresentationLearning #InterpretableML #Convolution Linking to them individually in the thread below 1/n
    Manolis Kellis @manoliskellis
    Interested in #DeepLearning for #Genomics and the #LifeSciences? Our @MIT course starts again today. Recommend your favorite papers and topics to cover, and we *will* cover them! Watch our 2020 lectures with @DavidKGifford at and
    15186673
  18. Zijian Dong✓@ZijianDD
    🧶 Introduce our @CVPR 2025 paper on representation learning for Deep Imbalanced Regression! Novel geometric constraints—Enveloping and Homogeneity—that encourage uniformity like yarn wrapped uniformly around a ball. Paper: Code:
    Download
    22769.9K
  19. Zijian Dong✓@ZijianDD
    @ZijianDD
    📊 Experiments on age estimation, text similarity, tabular data, and IOL confirm: ✅ Significant improvement on Few-shot & Medium-shot regions ✅ State-of-the-art results across multiple datasets
    102192
  21. Federico Pernici@FedPernici
    🔄 Retweeting our discussion on #CompatibleRepresentation. The ArXiv version of the revised CoReS paper is now available: Soon a brief overview! #RepresentationLearning #NeuralCollapse #DeepLearning #MachineLearning #TPAMI
    Federico Pernici @FedPernici
    🧩Three intriguing concepts in the world of #DeepLearning: #BackwardCompatibility, #StationaryRepresentation, and #NeuralCollapse. Their connection may hold the key to better deep learning models with possibly novel ways to use them (1/7)
    002567
  27. Matthias Niessner✓@MattNiessner
    Happening now! Come and join us if you want to learn more about #representationlearning, #transformers, #DiT, and so much more :)
    Matthias Niessner @MattNiessner
    Tomorrow in our TUM AI - Lecture Series we'll have none other than Saining Xie (@sainingxie), NYU He'll talk about "𝐓𝐡𝐞 𝐦𝐮𝐥𝐭𝐢𝐦𝐨𝐝𝐚𝐥 𝐟𝐮𝐭𝐮𝐫𝐞: Why visual representation still matters". Live stream: 5pm GMT+1 / 9am PST (Mon Mar 17th)
    044548.9K
  28. Matthias Niessner✓@MattNiessner
    Tomorrow in our TUM AI - Lecture Series we'll have none other than Saining Xie (@sainingxie), NYU He'll talk about "𝐓𝐡𝐞 𝐦𝐮𝐥𝐭𝐢𝐦𝐨𝐝𝐚𝐥 𝐟𝐮𝐭𝐮𝐫𝐞: Why visual representation still matters". Live stream: 5pm GMT+1 / 9am PST (Mon Mar 17th)
    Download
    75738258K
  31. Show more posts