A survey on modern trainable activation functions
May 02, 2020 Β· The Cartographer Β· π Neural Networks
"No code URL or promise found in abstract"
"Title-pattern auto-detect: A survey on modern trainable activation functions"
Evidence collected by the PWNC Scanner
Authors
Andrea Apicella, Francesco Donnarumma, Francesco IsgrΓ², Roberto Prevete
arXiv ID
2005.00817
Category
cs.LG: Machine Learning
Cross-listed
cs.NE,
stat.ML
Citations
485
Venue
Neural Networks
Last Checked
1 day ago
Abstract
In neural networks literature, there is a strong interest in identifying and defining activation functions which can improve neural network performance. In recent years there has been a renovated interest of the scientific community in investigating activation functions which can be trained during the learning process, usually referred to as "trainable", "learnable" or "adaptable" activation functions. They appear to lead to better network performance. Diverse and heterogeneous models of trainable activation function have been proposed in the literature. In this paper, we present a survey of these models. Starting from a discussion on the use of the term "activation function" in literature, we propose a taxonomy of trainable activation functions, highlight common and distinctive proprieties of recent and past models, and discuss main advantages and limitations of this type of approach. We show that many of the proposed approaches are equivalent to adding neuron layers which use fixed (non-trainable) activation functions and some simple local rule that constraints the corresponding weight layers.
Community Contributions
Found the code? Know the venue? Think something is wrong? Let us know!
π Similar Papers
In the same crypt β Machine Learning
ποΈ
ποΈ
Transcended
ποΈ
ποΈ
Transcended
Continuous control with deep reinforcement learning
π
π
Old Age
Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
π
π
Old Age
Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor
π
π
Old Age
SGDR: Stochastic Gradient Descent with Warm Restarts
ποΈ
ποΈ
Transcended