CORe50
PulseAugur coverage of CORe50 — every cluster mentioning CORe50 across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
New method preserves representation in domain-incremental learning
Researchers have developed a novel method for domain-incremental learning (DIL) that addresses the issue of performance degradation on previously learned domains when adapting to new ones. This approach tracks individua…
-
New SynGAP framework mimics biological metaplasticity for continual learning
Researchers have developed SynGAP, a novel continual learning framework that mimics biological metaplasticity to prevent catastrophic forgetting in artificial neural networks. Unlike existing methods that require task l…
-
New BPG Framework Enhances Domain Incremental Learning in Neural Networks
Researchers have developed BPG, a new framework for domain incremental learning (DIL) designed to improve how deep neural networks adapt to new data distributions without forgetting previous knowledge. BPG consists of t…
-
New CIFNet framework solves Class-Incremental Learning analytically
Researchers have developed CIFNet, a novel framework for Class-Incremental Learning (CIL) that bypasses traditional gradient-based optimization. By treating CIL as a sequence of deterministic, closed-form classifier ada…
-
HydraCIL offers efficient class-incremental learning for edge devices
Researchers have introduced HydraCIL, a novel approach to class-incremental learning designed for resource-constrained environments like embedded systems. This method decouples feature extraction from classifier trainin…
-
New framework corrects target shift in online learning systems
Researchers have developed a new framework to analyze and improve online learning systems that encounter distributional shifts. Their work, focusing on kernel regression, reveals that online learning effectively uses sh…
-
HEDP framework uses energy-distance prompts for domain incremental learning
Researchers have introduced HEDP, a new framework for domain incremental learning that aims to prevent performance degradation when models adapt to new data. The framework utilizes a hybrid approach combining energy-bas…
-
ImageHD accelerator boosts on-device continual learning with Hyperdimensional Computing
Researchers have developed ImageHD, a novel FPGA accelerator designed for energy-efficient on-device continual learning of visual representations using Hyperdimensional Computing (HDC). This system addresses the limitat…