• Corpus ID: 220280147

Deep Neural Networks as the Semi-classical Limit of Quantum Neural Networks

@article{Marcian2020DeepNN,
  title={Deep Neural Networks as the Semi-classical Limit of Quantum Neural Networks},
  author={Antonino Marcian{\`o} and De-Wei Chen and Filippo Fabrocini and Chris Fields and Enrico Greco and Niels G. Gresnigt and Krid Jinklub and Matteo Lulli and Kostas Terzidis and Emanuele Zappal{\`a}},
  journal={arXiv: Disordered Systems and Neural Networks},
  year={2020}
}
Our work intends to show that: (1) Quantum Neural Networks (QNN) can be mapped onto spin-networks, with the consequence that the level of analysis of their operation can be carried out on the side of Topological Quantum Field Theories (TQFT); (2) Deep Neural Networks (DNN) are a subcase of QNN, in the sense that they emerge as the semiclassical limit of QNN; (3) a number of Machine Learning (ML) key-concepts can be rephrased by using the terminology of TQFT. Our framework provides as well a… 
2 Citations

References

SHOWING 1-10 OF 51 REFERENCES

Sharp Minima Can Generalize For Deep Nets

It is argued that most notions of flatness are problematic for deep models and can not be directly applied to explain generalization, and when focusing on deep networks with rectifier units, the particular geometry of parameter space induced by the inherent symmetries that these architectures exhibit is exploited.

Towards Understanding Generalization of Deep Learning: Perspective of Loss Landscapes

The underlying reasons why deep neural networks often generalize well are investigated, and it is shown that the characteristics the landscape of the loss function that explains the good generalization capability is the volume of basin of attraction of good minima.

Temperley-Lieb Recoupling Theory and Invariants of 3-Manifolds

I used to keep the entire spin networks literature in a small folder on my shelf. The recent explosion of interest in the subject has made this impossible; more is probably now written every week

Understanding Machine Learning - From Theory to Algorithms

The aim of this textbook is to introduce machine learning, and the algorithmic paradigms it offers, in a principled way in an advanced undergraduate or beginning graduate course.

arXiv

The innovative preprint repository, arXiv, was created in the early 1990s to improve access to scientific research. arXiv contains millions of Open Access articles in physics, mathematics, computer

Conf

  • Ser. 314, 012006
  • 2011

Classical and Quantum Gravity 22

  • 1739
  • 2005

Generalization of knowledge : multidisciplinary perspectives

This chapter discusses Cognitive Neuroscience Approaches to Generalization, which focuses on the development of approaches to generalization in the context of human-Computer Interaction research.

Proceedings of the Royal Society (London)

  • Economics
    Science
  • 1906
...