Skip to content
conftrace_

theoretical insights

3 papers

Explore in graph

Co-occurring keywords · papers mentioning both

backdoor poisoning attacks (1) effectiveness (1) two-layer convolutional neural network (1) empirical results (1) optimal merging coefficient (1) multi-view data model (1) two-layer cnn (1) dirty-label attack (1) bayesian continual learning (1) stability-plasticity trade-off (1)

Papers

BECAME: Bayesian Continual Learning with Adaptive Model Merging ICML 2025 Towards Understanding Catastrophic Forgetting in Two-layer Convolutional Neural Networks ICML 2025 A Theoretical Analysis of Backdoor Poisoning Attacks in Convolutional Neural Networks ICML 2024