Error-Driven Incremental Learning in Deep Convolutional Neural Network for Large-Scale Image Classification

By Tianjun Xiao, Jiaxing Zhang, Kuiyuan Yang, Yuxin Peng, Zheng Zhang. 2014

256 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

ALICE: Active Learning with Contrastive Natural Language Explanations (2020, background): “Xiao et al. (2014) propose to expand the model based on category hierarchy for incremental learning.”
Recent advances in convolutional neural networks (2015, methodology): “[96] proposes a training method that grows a network”
iCaRL: Incremental Classifier and Representation Learning (2016, background): “[40] learn a tree-structured model that grows incrementally as more classes are observed.”
Rectification-based Knowledge Retention for Continual Learning (2021, methodology): “Dynamic network-based methods [33, 8, 3, 28, 15] use network expansions/modifications for training new tasks.”
Continual Semantic Segmentation via Repulsion-Attraction of Sparse and Disentangled Latent Representations (2021, background): “This phenomenon has been widely studied in image classification task and most of the current techniques fall into the following categories [10, 48]: regularization approaches [5, 32, 73, 13, 36], dynamic architectures [69, 64, 35], parameter isolation [17, 53, 40] and replay-based methods [66, 46, 5”
Knowledge Distillation for Incremental Learning in Semantic Segmentation (2019, methodology): “Focusing on deep neural networks, some methods [29, 30] exploit architectures which grow over time as a tree structure in a hierarchical manner as new classes are observed.”
Compacting, Picking and Growing for Unforgetting Continual Learning (2019, background): “To exactly avoid forgetting, a promising way is to keep the old-task weights already learned [38, 47] and enlarge the network by adding nodes or weights for training new tasks.”
Incremental Learning Techniques for Semantic Segmentation (2019, methodology): “Some methods [16, 27, 28, 36] exploit network architectures which grow during the training process.”
Large Scale Incremental Learning (2019, methodology): “[28] proposed a training method that grows a network hierarchically as new training data are added.”
End-to-End Incremental Learning (2018, background): “None of the existing approaches for incremental learning [4, 9, 13, 14, 16, 23, 26, 28, 30, 32, 34, 37] satisfy all these constraints.”
Class-Incremental Learning: Survey and Performance Evaluation on Image Classification (2020, background): “In [63], similar classes are grouped together expanding hierarchically, at the cost of an expensive training procedure and a rigid architecture.”
Incremental Classifier Learning with Generative Adversarial Networks (2018, methodology): “[25] proposed a training method that grows a network hierarchically as new training data are added.”
Sentence Embedding Alignment for Lifelong Relation Extraction (2019, background): “…with Dynamic Model Architecture There is another related direction on dynamically changing the model structure (i.e., adding new modules) in order to learn the new task without interfering learned knowledge for previous tasks, such as (Xiao et al., 2014; Rusu et al., 2016; Fernando et al., 2017).”
Revisiting Distillation and Incremental Classifier Learning (2018, background): “Xiao [26] proposed networks that grew as new classes were added to avoid changing weights important for older tasks.”
Incremental Learning for Semantic Segmentation of Large-Scale Remote Sensing Data (2018, background): “In [17], a tree-structured model, which grows hierarchically, is proposed.”
A Strategy for an Uncompromising Incremental Learner (2017, background): “A tree of deep convolutional networks (DCNN) for incremental learning was proposed by Xiao et al. [34].”
An Adaptive Random Path Selection Approach for Incremental Learning. (2019, background): “The existing adaptive network architectures come with their respective limitations e.g., [5]’s complexity grows with the tasks, [20] has an expensive training procedure and a somewhat rigid architecture and [21] does not allow incrementally learning new classes due to a detached output layer and…”