This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING…

Subfields: Model Architectures, Optimization & Training

Citations in the corpus

FearNet: Brain-Inspired Model for Incremental Learning (2017, background): “The Fixed Expansion Layer (FEL) model mitigates catastrophic forgetting by using sparse updates (Coop et al., 2013).”
Measuring Catastrophic Forgetting in Neural Networks (2017, methodology): “The Fixed Expansion Layer (FEL) model creates sparse representations by fixing the network’s weights and specifying neuron triggering conditions (Coop, Mishtal, and Arel 2013).”
Memory Efficient Experience Replay for Streaming Learning (2018, background): “Sparse-coding methods reduce the chance of new representations interfering with older representations [16].”
Lifelong Machine Learning with Deep Streaming Linear Discriminant Analysis (2019, background): “…regularizing weights to remain close to their previous values [3, 10, 15, 32, 45, 47, 52], promoting sparse weight updates to mitigate interference [13], and ensembling multiple classiﬁers [18], but recently, models that incorporate rehearsal (i.e., replay) have demonstrated the most success [9,…”
Graph-Based Continual Learning (2020, background): “First, the stochasticity induced by the Bernoulli random variables allows us to output multiple predictions and average these predictions, and such ensemble techniques have been quite successful in continual learning settings (Coop et al., 2013; Fernando et al., 2017).”
Adaptive Deep Models for Incremental Learning: Considering Capacity Scalability and Sustainability (2019, background): “parse representation [8], readers can refer to the introduction for further information [18].”
Cost-Effective Incremental Deep Model: Matching Model Capacity With the Least Sampling (2023, unknown): “getting by learning sparse representation [36].”
What and Where: Learn to Plug Adapters via NAS for Multidomain Learning (2020, background): “Various adapter modules are proposed to gain an acceptable performance, overcoming the “catastrophic forgetting” [13]–[15] problem.”
Neuron Clustering for Mitigating Catastrophic Forgetting in Supervised and Reinforcement Learning (2015, methodology): “The topic has grown in interest recently [15] [16] [17].”
Multiobjective Deep Belief Networks Ensemble for Remaining Useful Life Estimation in Prognostics (2017, background): “Many works [36]–[39] have demonstrated that a single method can seldom perform consistently well across a variety of applications.”
Overcoming Long-Term Catastrophic Forgetting Through Adversarial Neural Pruning and Synaptic Consolidation (2019, background): “[6] attempted to reserve task-specific structures for single tasks, such as layers or modules.”
Incremental Concept Learning via Online Generative Memory Recall (2019, background): “Fixed expansion layer(FEL) network[31] mitigates the forgetting by selectively update neural weights in the network rather than expanding the neural weights during continual learning.”
Incremental Learning Using Conditional Adversarial Networks (2019, background): “Among them, there is one kind that focuses on regularization of model parameters, such as [7], [19], and [5].”