SelfieBoost: A Boosting Algorithm for Deep Learning
By Shai Shalev-Shwartz. 2014
18 citations recorded by the source corpus.
We describe and analyze a new boosting algorithm for deep learning called SelfieBoost. Unlike other boosting algorithms, like AdaBoost, which construct ensembles of classifiers, SelfieBoost boosts the accuracy of a single network. We prove a $\log(1/\epsilon)$ convergence rate for SelfieBoost under some "SGD success" assumption which seems to hold in practice.
Subfields: Artificial Intelligence
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Sergey Levine
- Yoshua Bengio
- Razvan Pascanu
- Pieter Abbeel
- Ruslan Salakhutdinov
- Mohit Bansal
- Kyunghyun Cho
- Chelsea Finn
Citations in the corpus
- Policy Distillation (2015, methodology): “It also shows merit as an optimization method that acts to stabilize learning over large datasets or in dynamic domains (Shalev-Shwartz, 2014).”
- Learning Deep ResNet Blocks Sequentially using Boosting Theory (2017, background): “Shalev-Shwartz (Shalev-Shwartz, 2014) proposes a selfieBoost which boosts the accuracy of a single network.”