Incremental learning algorithms and applications

By Alexander Rainer Tassilo Gepperth, Barbara Hammer. 2016

333 citations recorded by the source corpus.

How researchers see this paper

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

Citations in the corpus

Rainbow Memory: Continual Learning with a Memory of Diverse Samples (2021, background): “Class-incremental learning usually refers to a sequential learning paradigm with disjoint set of tasks [39, 4, 15].”
Continual Learning for Robotics (2019, methodology): “, 2018), catastrophic forgetting (French, 1999), or data distribution shifts (Gepperth and Hammer, 2016).”
A comprehensive, application-oriented study of catastrophic forgetting in DNNs (2019, background): “The ﬁeld of incremental learning is large, e.g., Parisi et al. (2018) and Gepperth & Hammer (2016).”
Marginal Replay vs Conditional Replay for Continual Learning (2018, methodology): “…a DNN model with new tasks (e.g., visual classes) over long time periods, while avoiding the abrupt degradation of previously learned abilities that is known under the term "catastrophic interference" or "catastrophic forgetting" (Pfülb et al., 2018; French, 1999; Gepperth and Hammer, 2016).”
Confidence Calibration for Incremental Learning (2020, methodology): “Among the strategies summarized in [1], we are particularly interested in continual learning with a memory of previous tasks and no task identity given at the inference, which is referred to as ‘incremental learning’ in [12].”
Incremental Detection of Remote Sensing Objects With Feature Pyramid and Knowledge Distillation (2020, background): “Such a problem belongs to incremental learning, which refers to the situation of continuous model adaptation based on a constantly arriving data stream [15].”
A New Knowledge Distillation for Incremental Object Detection (2019, background): “For example, it is necessary for an unmanned driving vehicle to detect new classes of traffic signs and lane lines [2].”
Incremental Lifelong Deep Learning for Autonomous Vehicles (2018, background): “Online machine learning methods have been developed to allow models to adapt continuously to a stream of new data, including interactive scenarios where training examples are provided based on human feedback over time [5].”
FFNB: Forgetting-Free Neural Blocks for Deep Continual Learning (2021, methodology): “,T10 correspond to classes [51-100] while T0 is the pretraining task involving classes [1-50]).”
Incremental on-line learning: A review and comparison of state of the art algorithms (2018, background): “[87] V. Losing, B. Hammer, H. Wersing, Knn classifier with self adjusting memory for heterogeneous concept drift, in: 2016 IEEE 16th International Conference on Data Mining (ICDM), 2016, pp. 291–300. doi: 10.1109/ICDM.2016.0040.”
Class Boundary Exemplar Selection Based Incremental Learning for Automatic Target Recognition (2020, methodology): “Recently, incremental learning gained more attention especially in learning from data streams, and many deﬁnitions of incremental learning are introduced in the literature [7], [19], [23], [28].”
Continuum: Simple Management of Complex Continual Learning Scenarios (2021, background): “This type of drift is called real concept drift [14, 15].”
Continual Learning in Neural Networks (2019, result): “Catastrophic forgetting has also been attested in other machine learning models [52, 86].”
Incremental learning with self-organizing maps (2017, background): “Although attempts at formal definition ([2], [1]) conflict in minor points, they seem to agree that incremental learning algorithms”
CILEA-NET: Curriculum-Based Incremental Learning Framework for Remote Sensing Image Classification (2021, background): “On the other hand, the network’s reactivity drastically decreases if the updates are made gradually for retaining the learned information [2].”
Online Continual Learning on Sequences (2020, background): “fashion from each Di over X × Y [15, 21].”
IncDet: In Defense of Elastic Weight Consolidation for Incremental Object Detection (2020, background): “Incremental learning has been of great interest for decades [24], [25] and has attracted much attention recently [26], [27].”