5 papers
Suppression or Deletion: A Restoration-Based Representation-Level Analysis of Machine Unlearning
Yurim Jang, Jaeung Lee, Dohyun Kim +2
As pretrained models are increasingly shared on the web, ensuring that models can forget or delete sensitive, copyrighted, or private information upon request has become crucial. M…
Subnet-Aware Dynamic Supernet Training for Neural Architecture Search
Jeimin Jeon, Youngmin Oh, Junghyup Lee +4
N-shot neural architecture search (NAS) exploits a supernet containing all candidate subnets for a given search space. The subnets are typically trained with a static training stra…
Toward INT4 Fixed-Point Training via Exploring Quantization Error for Gradients
Dohyung Kim, Junghyup Lee, Jeimin Jeon +2
Network quantization generally converts full-precision weights and/or activations into low-bit fixed-point values in order to accelerate an inference process. Recent approaches to…
Scheduling Weight Transitions for Quantization-Aware Training
Junghyup Lee, Jeimin Jeon, Dohyung Kim +1
Quantization-aware training (QAT) simulates a quantization process during training to lower bit-precision of weights/activations. It learns quantized weights indirectly by updating…
Instance-Aware Group Quantization for Vision Transformers
Jaehyeon Moon, Dohyung Kim, Junyong Cheon +1
Post-training quantization (PTQ) is an efficient model compression technique that quantizes a pretrained full-precision model using only a small calibration set of unlabeled sample…