Фильтр публикаций


The Abelian sandpile and related models

https://arxiv.org/pdf/cond-mat/9808047


book1.pdf
354.0Кб
Lattice theory


221113854v5_250110_022829.pdf
4.5Мб
ComCLIP: Training-Free Compositional Image and Text Matching


200615055v2_250110_021932.pdf
3.2Мб
Compositional generalization: Object-Centric Learning with Slot Attention




2001.08361v1.pdf
2.4Мб
Scaling Laws for Neural Language Models


2106.04496v3.pdf
1.1Мб
Towards a Theoretical Framework of
Out-of-Distribution Generalization


1610.02136v3.pdf
206.3Кб
A BASELINE FOR DETECTING MISCLASSIFIED AND
OUT-OF-DISTRIBUTION EXAMPLES
IN NEURAL NETWORKS


TrojanNet_Embedding_Hidden_Trojan_Horse_Models_in_.pdf
2.3Мб
NP-completeness proof of trajanNet detection






Scanning Trojaned Models Using Out-of-Distribution
Samples

https://openreview.net/pdf?id=m296WJXyzQ


171004019v2_241214_150725.pdf
3.6Мб
An introduction to Topological Data Analysis: fundamentaland practical aspects for data scientists


Zhang,yujie_241214_150408.pdf
634.8Кб
PERSISTENT HOMOLOGY AND SPARSE VIETORIS-RIPSFILTRATION


Репост из: اتاق ریاضی علامه حلی
proof.pdf
14.7Мб
کتاب اثبات
نویسنده: مارتین ایگنر | گونتر تسیگلر
مترجم: سیامک کاظمی

@HelliMathroom

#کتاب


10.1109_mlsp58920.2024.10734765_cbey.pdf
772.2Кб
IMPROVED ACTIVATION CLIPPING FOR UNIVERSAL BACKDOOR MITIGATION ANDTEST-TIME DETECTION

Thanks to mehrzad


231202673v1_241214_123737.pdf
2.6Мб
Robust Backdoor Detection for Deep Learning via Topological Evolution Dynamics


Видео недоступно для предпросмотра
Смотреть в Telegram


survey.pdf
1.5Мб
I've extracted previous information about backdoor attacks using this survey


Almost all of the backdoor defense methods were bypassed by this two attacks!

Показано 20 последних публикаций.