Black holes and the loss landscape in machine learning
From MaRDI portal
Publication:6061784
DOI10.1007/jhep10(2023)107arXiv2306.14817OpenAlexW4387742192MaRDI QIDQ6061784
Swapnamay Mondal, Taniya Mandal, Pranav Kumar
Publication date: 8 December 2023
Published in: Journal of High Energy Physics (Search for Journal in Brave)
Full work available at URL: https://arxiv.org/abs/2306.14817
Cites Work
- Unnamed Item
- Unnamed Item
- BPS state counting in \(N=8\) supersymmetric string theory for pure D-brane configurations
- Arithmetic of \( \mathcal{N} = 8 \) black holes
- On the positivity of black hole degeneracies in string theory
- Particle creation by black holes
- Elliptic genera and quantum field theory
- How do black holes predict the sign of the Fourier coefficients of Siegel modular forms?
- Supersymmetric index from black hole entropy
- Supersymmetry and Morse theory
- \(R^2\) corrections and non-perturbative dualities of \(N=4\) string ground states
- Black hole entropy in M-theory
- Microscopic origin of the Bekenstein-Hawking entropy
- Evolving neural networks with genetic algorithms to study the string landscape
- Machine learning in the string landscape
- Deep learning in the heterotic orbifold landscape
- Scaling BPS solutions and pure-Higgs states
- Moduli-dependent Calabi-Yau and SU(3)-structure metrics from machine learning
- Scaling black holes and modularity
- Gradient descent optimizes over-parameterized deep ReLU networks
- Data science applications to string theory
- Machine learning Calabi-Yau four-folds
- Machine learning Lie structures \& applications to physics
- Getting CICY high
- Complexity of random smooth functions on the high-dimensional sphere
- The four laws of black hole mechanics
- Replica symmetry breaking condition exposed by random matrix calculation of landscape complexity
- Neural network approximations for Calabi-Yau metrics
- Some NP-complete problems in quadratic and nonlinear programming
- Mean-field theory for a spin-glass model of neural networks: TAP free energy and the paramagnetic to spin-glass transition
- Training a Single Sigmoidal Neuron Is Hard
- Random Matrices and Complexity of Spin Glasses
- Role of initial conditions in the dynamics of quantum glassy systems
- Deep learning the holographic black hole with charge
- A mathematical theory of semantic development in deep neural networks
- Black Holes and Entropy
- Comparing dynamics: deep neural networks versus glassy systems
- Entropy-SGD: biasing gradient descent into wide valleys
- \(F^4\) terms in \(N=4\) string vacua
- Connecting Dualities and Machine Learning
This page was built for publication: Black holes and the loss landscape in machine learning
