% IMPORTANT: The following is UTF-8 encoded.  This means that in the presence
% of non-ASCII characters, it will not work with BibTeX 0.99 or older.
% Instead, you should use an up-to-date BibTeX implementation like “bibtex8” or
% “biber”.

@PROCEEDINGS{Draganov:1047276,
      author       = {Draganov, Andrew Alexander and Sharvaree, Vadgama and
                      Sebastian, Damrich and Böhm, Jan Niklas and Maes, Lucas and
                      Kobak, Dmitry and Bekkers, Erik},
      title        = {{O}n the {I}mportance of {E}mbedding {N}orms in
                      {S}elf-{S}upervised {L}earning; {PMLR} 267, 2025},
      volume       = {267},
      reportid     = {FZJ-2025-04197},
      year         = {2025},
      note         = {$https://icml.cc/media/PosterPDFs/ICML\%202025/45610.png?t=1751969010.7689042$},
      abstract     = {AbstractSelf-supervised learning (SSL) allows training data
                      representations without a supervised signal and has become
                      an important paradigm in machine learning. Most SSL methods
                      employ the cosine similarity between embedding vectors and
                      hence effectively embed data on a hypersphere. While this
                      seemingly implies that embedding norms cannot play any role
                      in SSL, a few recent works have suggested that embedding
                      norms have properties related to network convergence and
                      confidence. In this paper, we resolve this apparent
                      contradiction and systematically establish the embedding
                      norm’s role in SSL training. Using theoretical analysis,
                      simulations, and experiments, we show that embedding norms
                      (i) govern SSL convergence rates and (ii) encode network
                      confidence, with smaller norms corresponding to unexpected
                      samples. Additionally, we show that manipulating embedding
                      norms can have large effects on convergence speed. Our
                      findings demonstrate that SSL embedding norms are integral
                      to understanding and optimizing network behavior.},
      month         = {Jul},
      date          = {2025-07-13},
      organization  = {42 nd International Conference on
                       Machine Learning, Vancouver, Canada.
                       (Canada), 13 Jul 2025 - 19 Jul 2025},
      cin          = {IAS-8},
      cid          = {I:(DE-Juel1)IAS-8-20210421},
      pnm          = {5112 - Cross-Domain Algorithms, Tools, Methods Labs (ATMLs)
                      and Research Groups (POF4-511)},
      pid          = {G:(DE-HGF)POF4-5112},
      typ          = {PUB:(DE-HGF)26},
      url          = {https://juser.fz-juelich.de/record/1047276},
}