Deschide articolul original într-un tab nou Schimbarea acestei opțiuni se aplică imediat și reîncarcă pagina.
LECTURA
Înapoi la articole →

argmax(blog)

argmax.blog/index.xml
Vizitează site-ul

Cel mai recent articol

Ultima verificare

Urmăritori 0 urmăritori
Articole: 14

  1. A quantitative measure of mode-seeking vs. mode-covering preferences

    (argmax(blog))

    Another view

  2. Stein discrepancy is mode-seeking

    (argmax(blog))

    Stein discrepancy is more mode-seeking than reverse KL

  3. Sampler evaluation: No mode-covering without importance sampling

    (argmax(blog))

    An impossibility result

  4. Defining mode-covering behavior with gradient analysis

    (argmax(blog))

    What does mode-seeking vs. mode-covering behavior in distributional divergences really mean?

  5. Outranking leaderboard models on test sets, with only their predictions

    (argmax(blog))

    A brainteaser

  6. On diversity and many-model ensembling: AI government & AI-augmented public goods funding

    (argmax(blog))

    When AI agents outnumber voters, how should we elect an AI government?

  7. The deep retrieval + remixing hypothesis

    (argmax(blog))

    How do modern deep generative models produce high-quality outputs?

  8. What is data? The classical and postmodern views

    (argmax(blog))

    Is data sacred?

  9. On Creating AGI within Capitalism -- The Greatest Threat to AGI is ANI.

    (argmax(blog))

    A speculative argument

  10. Implicit regularization of SGD in a linear model

    (argmax(blog))

    A brief mathematical result

  11. Treating high-dimensional prediction/generation as optimization

    (argmax(blog))

    Exploring connections between diffusion models, alphafold's recycling, structured prediction energy networks, and manifold learning.

  12. Primer on Score-based Generative Models

    (argmax(blog))

    An introduction

  13. Implicit Differentiation through Equilibria

    (argmax(blog))

    An introduction.

  14. On KL Divergence in Discrete Spaces

    (argmax(blog))

    KL behaves differently in discrete spaces than in continuous spaces.