Overcoming catastrophic forgetting in neural networks

By OctoMiao. 2016

144 citations recorded by the source corpus.

Subfields: Meta & Continual Learning

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

1. Zixuan Ke
2. Bin Liu
3. Gyuhak Kim
4. Sung Ju Hwang
5. Jaehong Yoon
6. Yarin Gal
7. Sebastian Farquhar
8. Michael C. Mozer

Citations in the corpus

• Overcoming Catastrophic Interference by Conceptors (2017, methodology): “Another task on which we tested CAB is the permuted MNIST experiment [Goodfellow et al., 2014; Kirkpatrick et al., 2017; Lee et al., 2017; Srivastava et al., 2013a], where a sequence of pattern recognition tasks are created from the MNIST dataset [LeCun et al., 1998].”
• A Unifying Bayesian View of Continual Learning (2019, methodology): “Although they are not formulated using BNNs, Elastic Weight Consolidation (EWC) [Kirkpatrick et al., 2017], Synaptic Intelligence (SI) [Zenke et al., 2017], Riemanian Walk [Chaudhry et al., 2018], and the extension of EWC offered by Ritter et al. [2018] all rely on a related Bayesian justification…”
• A Multi-Head Model for Continual Learning via Out-of-Distribution Replay (2022, background): “Using regularizations (Kirkpatrick et al., 2017) is a popular approach.”