Bayesian Gradient Descent: Online Variational Bayes Learning with Increased Robustness to Catastrophic Forgetting and Weight Pruning

18 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.

Task Agnostic Continual Learning via Meta Learning (2019, methodology): “We show that our method is capable of forward transfer by comparing it to Reptile, BGD and SGD on the sine regression tasks commonly used in meta learning literature.”
Incremental Learning for Semantic Segmentation of Large-Scale Remote Sensing Data (2018, methodology): “When the data come sequentially, the works explained in [37]–[39] optimize parameters of the network by updating the posterior approximation by the Bayesian inference based methods.”
Partitioned Variational Inference: A unified framework encompassing federated and continual learning (2018, background): “…& Attias 2000 Sato 2001 Broderick et al. 2013 Bui et al. 2018 Winn et al. 2005 Archambeau & Ermis, 2015 Hinton and van Camp 1993 Models encompassed: conjugate low dimensional non-conjugate high dimensional non-conjugate Zeno et al. 2018 Figure 3: The local VI framework unifies prior work.”
Energy-Based Models for Continual Learning (2020, methodology): “We use the code of “continuous task-agnostic learning” proposed by (Zeno et al., 2018) to generate the training data stream.”