Reconciling meta-learning and continual learning with online mixtures of tasks

By Ghassen Jerfel, Erin Grant, T. Griffiths, Katherine A. Heller. 2018

113 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.

In Meta & Continual Learning

Citations in the corpus

A Comprehensive Survey of Continual Learning: Theory, Method and Application (2023, methodology): “CN-DPM [80] adopts Dirichlet process mixture models to construct a growing number of neural experts, while a concurrent work [269] derives such mixture models from a probabilistic meta-learner.”
Balancing Stability and Plasticity through Advanced Null Space in Continual Learning (2022, methodology): “These method can be roughly divided into four categories: architecture-based methods expand the network or allocate new neurons for new tasks [13,28,42,60]; replayedbased methods interleave old data with current data by storing historical data in a buffer or generating virtual old data [4,8,12,39,40”
CoMPS: Continual Meta Policy Search (2021, background): “Recent work has made progress toward using supervised meta-learning in an online setting (Ren et al., 2018; Finn et al., 2019; Khodak et al., 2019; Jerfel et al., 2019; Grant et al., 2019; Zhuang et al., 2019; Antoniou et al., 2020; Wallingford et al., 2020; Yu et al., 2020a; Yao et al., 2020).”
Continual Learning via Local Module Composition (2021, methodology): “Continual learning methods typically address the problem of forgetting through parameter regularization [46, 70, 99], replay [89, 78, 3, 73, 54, 12, 97, 39, 81, 13] or dynamic architectures (and MoEs) [83, 87, 52, 85, 53, 41].”
Kernel Continual Learning (2021, background): “Dynamic architectures (Rusu et al., 2016; Yoon et al., 2018; Jerfel et al., 2019; Li et al., 2019) allocate a subset of the model parameters for each task.”
Learngene: From Open-World to Your Learning Task (2021, background): “Although meta continual learning (Yao et al. 2020; Gidaris and Komodakis 2018; Jerfel et al. 2019; Javed and White 2019; Gupta, Yadav, and Paull 2020) can solve the sequential task problem by recording the efﬁciency task buffer, it still cannot prevent data-interaction.”
Learn Continually, Generalize Rapidly: Lifelong Knowledge Accumulation for Few-shot Learning (2021, background): “The problem setup of Jerfel et al. (2019) is most related to ours which learns to perform few-shot learning on new tasks better, but is only studied for image classification tasks with much smaller number tasks.”
Linear Mode Connectivity in Multitask and Continual Learning (2020, methodology): “In this sense it connects to topics such as MetaLearning (Beaulieu et al., 2020; Jerfel et al., 2019; He & Jaeger, 2018; Riemer et al., 2018), Few-Shot Learning (Wen et al., 2018; Gidaris & Komodakis, 2018), Multi-task and Transfer Learning (He & Jaeger, 2018; Jerfel et al., 2019).”
Incremental Meta-Learning via Indirect Discriminant Alignment (2020, background): “[15, 11] on the other hand, tackled a continual/online meta-learning setup where an explicit delineation between different tasks is not available, whereas in our experimental setup we are primarily trying to solve new classification tasks with clear task boundaries.”
A Neural Dirichlet Process Mixture Model for Task-Free Continual Learning (2020, background): “eled by each component can be decomposed as p(x;yjz) = p(yjx;z)p(xjz) (Rasmussen & Ghahramani, 2002; Shahbaba & Neal, 2009). DPM in Related Fields. Recent works of Nagabandi et al. (2019) and Jerfel et al. (2019) exploit the DPM framework to add new components without supervision in the meta”
Orthogonal Gradient Descent for Continual Learning (2019, methodology): “Meta learning algorithms use a data-driven inductive bias to enhance learning new tasks (Jerfel et al., 2019; He and Jaeger, 2018; Vuorio et al., 2018; Al-Shedivat et al., 2017; Riemer et al., 2018).”
Continual Few-Shot Learning for Text Classification (2021, background): “Online learning algorithms learn to update models from data streams sequentially, where the task is the same but can exhibit concept drift (new patterns) (Zinkevich, 2003; Crammer et al., 2006; Sahoo et al., 2018; Jerfel et al., 2019; Javed and White, 2019”
Online Fast Adaptation and Knowledge Accumulation: a New Approach to Continual Learning (2020, methodology): “Dirchlet-based meta learning (DBML) (Jerfel et al., 2019) uses a Dirchlet mixture model to infer the task identities sequentially.”
SOLA: Continual Learning with Second-Order Loss Approximation (2020, methodology): “ercome catastrophic forgetting [8]. The expansion based methods allocate new neurons or layers or modules to accommodate new tasks while utilizing the shared representation learned from previous ones [40, 45, 46, 25, 17]. Although being a very natural approach the mechanism of dynamic expansion can ”
Graph-Based Continual Learning (2020, neutral): “Recently, non-parametric Bayesian approaches use Dirichlet process mixture models to expand a set of neural networks in a principled way (Jerfel et al., 2019; Lee et al., 2020).”
Online Fast Adaptation and Knowledge Accumulation (OSAKA): a New Approach to Continual Learning (2020, methodology): “Dirchlet-based meta learning (DBML) [30] uses a Dirchlet mixture model to infer the task identities sequentially.”
TaskNorm: Rethinking Batch Normalization for Meta-Learning (2020, background): “In particular, meta-learning systems have been demonstrated to excel at complex learning tasks such as few-shot (Snell et al., 2017; Finn et al., 2017) and continual learning (Nagabandi et al., 2019; Requeima et al., 2019a; Jerfel et al., 2019).”
Lifelong Infinite Mixture Model Based on Knowledge-Driven Dirichlet Process (2021, background): “structure [10, 14, 47] in which each expert is built on the top of a shared module and only a single expert is updated during the training.”