Corpus ID: 22909921

Deep Learning Works in Practice. But Does it Work in Theory?

@article{Hoang2018DeepLW,
  title={Deep Learning Works in Practice. But Does it Work in Theory?},
  author={L{\^e} Nguy{\^e}n Hoang and Rachid Guerraoui},
  journal={ArXiv},
  year={2018},
  volume={abs/1801.10437}
}
Deep learning relies on a very specific kind of neural networks: those superposing several neural layers. In the last few years, deep learning achieved major breakthroughs in many tasks such as image analysis, speech recognition, natural language processing, and so on. Yet, there is no theoretical explanation of this success. In particular, it is not clear why the deeper the network, the better it actually performs. We argue that the explanation is intimately connected to a key feature of the… Expand
A Novel Approach in Determining Neural Networks Architecture to Classify Data With Large Number of Attributes
TLDR
A new approach is proposed to determine the neural networks architecture especially in the form of Multi-layer Perceptron (MLP) which will later be used as a machine learning method to classify data with large number of attribute. Expand
Separating the Structural Components of Maize for Field Phenotyping Using Terrestrial LiDAR Data and Deep Convolutional Neural Networks
TLDR
The proposed voxel-based convolutional neural network demonstrated LiDAR’s ability to separate structural components for crop phenotyping using deep learning, which can be useful for other fields. Expand

References

SHOWING 1-10 OF 30 REFERENCES
Why Does Deep and Cheap Learning Work So Well?
TLDR
It is argued that when the statistical process generating the data is of a certain hierarchical form prevalent in physics and machine learning, a deep neural network can be more efficient than a shallow one. Expand
Deep Networks with Stochastic Depth
TLDR
Stochastic depth is proposed, a training procedure that enables the seemingly contradictory setup to train short networks and use deep networks at test time and reduces training time substantially and improves the test error significantly on almost all data sets that were used for evaluation. Expand
The Unreasonable Effectiveness of Deep Learning
We show how well known rules of back propagation arise from a weighted combination of finite automata. By redefining a finite automata as a predictor we combine the set of all k-state finite automataExpand
On the Expressive Power of Deep Neural Networks
We propose a new approach to the problem of neural network expressivity, which seeks to characterize how structural properties of a neural network family affect the functions it is able to compute.Expand
On the Expressive Power of Deep Architectures
TLDR
Some of the theoretical motivations for deep architectures, as well as some of their practical successes, are reviewed, and directions of investigations to address some of the remaining challenges are proposed. Expand
Exponential expressivity in deep neural networks through transient chaos
TLDR
The theoretical analysis of the expressive power of deep networks broadly applies to arbitrary nonlinearities, and provides a quantitative underpinning for previously abstract notions about the geometry of deep functions. Expand
Representation Benefits of Deep Feedforward Networks
This note provides a family of classification problems, indexed by a positive integer $k$, where all shallow networks with fewer than exponentially (in $k$) many nodes exhibit error at least $1/6$,Expand
The Power of Depth for Feedforward Neural Networks
TLDR
It is shown that there is a simple (approximately radial) function on $\reals^d$, expressible by a small 3-layer feedforward neural networks, which cannot be approximated by any 2-layer network, unless its width is exponential in the dimension. Expand
Benefits of Depth in Neural Networks
TLDR
This result is proved here for a class of nodes termed "semi-algebraic gates" which includes the common choices of ReLU, maximum, indicator, and piecewise polynomial functions, therefore establishing benefits of depth against not just standard networks with ReLU gates, but also convolutional networks with reLU and maximization gates, sum-product networks, and boosted decision trees. Expand
Logical depth and physical complexity
Some mathematical and natural objects (a random sequence, a sequence of zeros, a perfect crystal, a gas) are intuitively trivial, while others (e.g. the human body, the digits of π) contain internalExpand
...
1
2
3
...