Fetril: Feature translation for exemplar-free class-incremental learning
Exemplar-free class-incremental learning is very challenging due to the negative effect of
catastrophic forgetting. A balance between stability and plasticity of the incremental process …
catastrophic forgetting. A balance between stability and plasticity of the incremental process …
Fecam: Exploiting the heterogeneity of class distributions in exemplar-free continual learning
Exemplar-free class-incremental learning (CIL) poses several challenges since it prohibits
the rehearsal of data from previous tasks and thus suffers from catastrophic forgetting …
the rehearsal of data from previous tasks and thus suffers from catastrophic forgetting …
Self-supervision can be a good few-shot learner
Existing few-shot learning (FSL) methods rely on training with a large labeled dataset, which
prevents them from leveraging abundant unlabeled data. From an information-theoretic …
prevents them from leveraging abundant unlabeled data. From an information-theoretic …
Few-shot class incremental learning leveraging self-supervised features
Abstract Few-Shot Class Incremental Learning (FSCIL) is a recently introduced Class
Incremental Learning (CIL) setting that operates under more constrained assumptions: only …
Incremental Learning (CIL) setting that operates under more constrained assumptions: only …
Variable few shot class incremental and open world learning
Prior work on few-shot class incremental learning has operated with an unnatural
assumption: the number of ways and number of shots are assumed to be known and fixed …
assumption: the number of ways and number of shots are assumed to be known and fixed …
A review of open-world learning and steps toward open-world learning without labels
In open-world learning, an agent starts with a set of known classes, detects, and manages
things that it does not know, and learns them over time from a non-stationary stream of data …
things that it does not know, and learns them over time from a non-stationary stream of data …
An open-world time-series sensing framework for embedded edge devices
A Bukhari, S Hosseinimotlagh… - 2022 IEEE 28th …, 2022 - ieeexplore.ieee.org
The rapid advancement of IoT technologies has generated much interest in the development
of learning-based sensing applications on embedded edge devices. However, these efforts …
of learning-based sensing applications on embedded edge devices. However, these efforts …
OpenSense: An Open-World Sensing Framework for Incremental Learning and Dynamic Sensor Scheduling on Embedded Edge Devices
A Bukhari, S Hosseinimotlagh… - IEEE Internet of Things …, 2024 - ieeexplore.ieee.org
Recent advances in Internet-of-Things (IoT) technologies have sparked significant interest
towards developing learning-based sensing applications on embedded edge devices …
towards developing learning-based sensing applications on embedded edge devices …
Five Roads to Open-Set Recognition
AR Dhamija - 2022 - search.proquest.com
When applying computer vision systems in the real world, the problem of detecting which
test samples do not belong to one of the semantic classes in a classifier's training set …
test samples do not belong to one of the semantic classes in a classifier's training set …
CTR: Contrastive Training Recognition Classifier for Few-Shot Open-World Recognition
N Dionelis, SA Tsaftaris… - 2022 26th International …, 2022 - ieeexplore.ieee.org
AI-enabled systems in security, autonomous systems, safety, and healthcare do not only
need to effectively detect Out-of-Distribution (OoD) samples, but also to recognize Objects of …
need to effectively detect Out-of-Distribution (OoD) samples, but also to recognize Objects of …