The paper introduces a continual learning framework combining synthetic sketch generation and trusted sample replay to enable a single model to perform multiple sketch biometric identification tasks.
icarl: Incremental classifier and representation learning,
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
verdicts
UNVERDICTED 5representative citing papers
ImageHD delivers up to 40.4x speedup and 383x energy efficiency for on-device continual learning of visual representations by using hyperdimensional computing and bounded exemplar management on an FPGA.
iGSP uses implicit gradient subspace projection in two phases to enable efficient continual adaptation of vision-language models, claiming SOTA accuracy with 42.7% fewer trainable parameters and 86.9% less total parameter growth.
RETROFIT enables continual learning for malware detection and binary summarization by retrospective-free parameter merging with low-rank sparse updates and confidence-guided arbitration, improving retention and generalization without historical data.
DualOpt decouples optimization by using real-time layer-wise weight decay for scratch training and weight rollback for fine-tuning to improve convergence, generalization, and reduce knowledge forgetting.
citing papers explorer
-
Bridging Data Trials and Task Barriers: A Unified Framework for Sketch Biometric Identification
The paper introduces a continual learning framework combining synthetic sketch generation and trusted sample replay to enable a single model to perform multiple sketch biometric identification tasks.
-
ImageHD: Energy-Efficient On-Device Continual Learning of Visual Representations via Hyperdimensional Computing
ImageHD delivers up to 40.4x speedup and 383x energy efficiency for on-device continual learning of visual representations by using hyperdimensional computing and bounded exemplar management on an FPGA.
-
iGSP:Implicit Gradient Subspace Projection for Efficient Continual Learning of Vision-Language Models
iGSP uses implicit gradient subspace projection in two phases to enable efficient continual adaptation of vision-language models, claiming SOTA accuracy with 42.7% fewer trainable parameters and 86.9% less total parameter growth.
-
Retrofit: Continual Learning with Controlled Forgetting for Binary Security Detection and Analysis
RETROFIT enables continual learning for malware detection and binary summarization by retrospective-free parameter merging with low-rank sparse updates and confidence-guided arbitration, improving retention and generalization without historical data.
-
Neural Network Optimization Reimagined: Decoupled Techniques for Scratch and Fine-Tuning
DualOpt decouples optimization by using real-time layer-wise weight decay for scratch training and weight rollback for fine-tuning to improve convergence, generalization, and reduce knowledge forgetting.