İçeriğe geç

Kanaka Rajan

@kanakarajanphd.bsky.social did:plc:nemx43ymm6toi4fqil4gqbev

Associate Professor at Harvard & Kempner Institute. Applying computational frameworks & machine learning to decode multi-scale neural processes. Marathoner. Rescue dog mom. https://www.rajanlab.com/

Bluesky’a katılma tarihi:

Takipçi
3.785
Takip edilen
248
Post
61
Sıralamaya girdiği gün
0

Özet

Kanaka Rajan (@kanakarajanphd.bsky.social), 11 Eylül 2023 tarihinde katılan bir Bluesky kullanıcısı. 3.785 takipçisi var, 248 hesabı takip ediyor ve 61 post paylaştı.

Sıralama geçmişi

Bu hesabın postlarının Debya günlük listesine girdiği en iyi günler.

Bu hesap henüz günlük sıralamaya girmedi.

Son beğendiği postlar

Bluesky ağındaki herkese açık beğeniler.

  • Ann Huang@annhuang42.bsky.social

    📄 Paper: arxiv.org/abs/2609.38081 🌐 Page: ann-huang-0.github.io/Hessian-null... 💻 Code: github.com/Ann-Huang-0/hessian-null-space-continuation w/ my awesome team 🌟 @neurostrow.bsky.social, @hansonlu.bsky.social @wtredman.bsky.social, @leokoz8.bsky.social, @kanakarajanphd.bsky.social (12/12)

    Postu göster
  • Ann Huang@annhuang42.bsky.social

    Overall, HNC finds diverse mechanisms with matched function around a single trained network. This opens the door to mechanistic analysis of solution spaces, to separating task demands from optimizer bias, and to a principled basis for reward auditing, model merging, editing and fine-tuning (11/12)

    Postu göster
  • Ann Huang@annhuang42.bsky.social

    📐 HNC also measures loss-landscape geometry. The Hessian captures the local geometry around one network; the null space walk tracks how it changes along the way. Bigger models have more flat directions, while harder tasks have fewer. The directions HNC follows get stiffer on harder tasks. (10/12)

    Postu göster
  • Ann Huang@annhuang42.bsky.social

    HNC can also uncover reward hacking! In the AI Safety Gridworld boat race, we start with policies that correctly follow the track. Preserving proxy reward (+3 for entering an arrow tile clockwise), HNC walks find the known exploit: hopping on and off one tile without net clockwise progress (9/12)

    Postu göster
  • Ann Huang@annhuang42.bsky.social

    Applied to RL, HNC searches for different policies while preserving a reward surrogate. An odor-tracking agent shifts from zigzagging across a plume to smoothly following its edge—a distinct navigation strategy at comparable return 🪰(8/12)

    Postu göster

Daha fazlası için üye girişi yap

Blokladığı hesaplar

Bluesky’da bloklar herkese açık kayıtlardır.

Bu hesap kimseyi bloklamıyor.