Comparing Incremental Learning Strategies for Convolutional Neural Networks

By Vincenzo Lomonaco, Davide Maltoni. 2016

23 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

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

1. Timothée Lesort
2. Joost van de Weijer
3. Vincenzo Lomonaco
4. David Filliat
5. Natalia Díaz Rodríguez
6. Davide Maltoni
7. Andrew D. Bagdanov
8. Bartłomiej Twardowski

In Computer Vision and Pattern Recognition

1. Joost van de Weijer
2. Timothée Lesort
3. Vincenzo Lomonaco
4. Andrew D. Bagdanov
5. Bartłomiej Twardowski
6. Xialei Liu
7. Marc Masana
8. David Filliat

Citations in the corpus

• Class-Incremental Learning: Survey and Performance Evaluation on Image Classification (2020, background): “Lomonaco and Maltoni [15] evaluate several strategies on weight initialization: random initialization, starting from a pretrained network and starting from a pretrained but only learning the classifier.”
• Continual Learning for Robotics (2019, background): “Initial progresses in this area have been proposed in (Thrun and Mitchell, 1995; Pasquale et al., 2015; Lomonaco and Maltoni, 2016; Camoriano et al., 2017; Lomonaco and Maltoni, 2017).”
• Online Continual Learning on Sequences (2020, background): “BigBrother dataset (see [40] [12]) is an interesting incremental learning setup in the face recognition domain but copyright restrictions do not allow the public distribution of the dataset.”
• Trojaning Attack on Neural Networks (2018, methodology): “We applied the incremental learning technique in [31], which does not require the original training data or the reverse engineered training data.”