RCL: Reliable Continual Learning for Unified Failure Detection
Deep neural networks are known to be overconfident for what they don't know in the wild
which is undesirable for decision-making in high-stakes applications. Despite quantities of …
which is undesirable for decision-making in high-stakes applications. Despite quantities of …
CrossMAE: Cross-Modality Masked Autoencoders for Region-Aware Audio-Visual Pre-Training
Learning joint and coordinated features across modalities is essential for many audio-visual
tasks. Existing pre-training methods primarily focus on global information neglecting fine …
tasks. Existing pre-training methods primarily focus on global information neglecting fine …
Flipped classroom: Aligning teacher attention with student in generalized category discovery
Recent advancements have shown promise in applying traditional Semi-Supervised
Learning strategies to the task of Generalized Category Discovery (GCD). Typically, this …
Learning strategies to the task of Generalized Category Discovery (GCD). Typically, this …
Interpreting pretext tasks for active learning: a reinforcement learning approach
As the amount of labeled data increases, the performance of deep neural networks tends to
improve. However, annotating a large volume of data can be expensive. Active learning …
improve. However, annotating a large volume of data can be expensive. Active learning …
Happy: A Debiased Learning Framework for Continual Generalized Category Discovery
Constantly discovering novel concepts is crucial in evolving environments. This paper
explores the underexplored task of Continual Generalized Category Discovery (C-GCD) …
explores the underexplored task of Continual Generalized Category Discovery (C-GCD) …
Deep Active Learning in the Open World
Machine learning models deployed in open-world scenarios often encounter unfamiliar
conditions and perform poorly in unanticipated situations. As AI systems advance and find …
conditions and perform poorly in unanticipated situations. As AI systems advance and find …
Uncertainty Quantification in Continual Open-World Learning
AI deployed in the real-world should be capable of autonomously adapting to novelties
encountered after deployment. Yet, in the field of continual learning, the reliance on novelty …
encountered after deployment. Yet, in the field of continual learning, the reliance on novelty …
Beyond Known Clusters: Probe New Prototypes for Efficient Generalized Class Discovery
Generalized Class Discovery (GCD) aims to dynamically assign labels to unlabelled data
partially based on knowledge learned from labelled data, where the unlabelled data may …
partially based on knowledge learned from labelled data, where the unlabelled data may …
Freeze and Cluster: A simple baseline for Rehearsal-Free Continual Category Discovery
X Yu, C Zhang, P Gu, X He - openreview.net
This paper addresses the problem of Rehearsal-Free Continual Category Discovery (RF-
CCD), which focuses on continuously identifying novel class by leveraging knowledge from …
CCD), which focuses on continuously identifying novel class by leveraging knowledge from …
Debiased Imbalanced Pseudo-Labeling for Generalized Category Discovery
Generalized Category Discovery (GCD) is a challenging task that aims to recognize seen
and novel categories within unlabeled data by leveraging labeled data. Designing a …
and novel categories within unlabeled data by leveraging labeled data. Designing a …