Priors for Infinite Networks
By Radford M. Neal. 1996
Subfields: Artificial Intelligence, Probabilistic & Bayesian Methods, Statistical Learning Theory
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
- Yoshua Bengio
- Ruslan Salakhutdinov
- Sergey Levine
- Quoc V. Le
- Mohit Bansal
- Percy Liang
- Jianfeng Gao
- Kyunghyun Cho
In Probabilistic & Bayesian Methods
- David M. Blei
- Zoubin Ghahramani
- Yoshua Bengio
- Yee Whye Teh
- Ruslan Salakhutdinov
- Michael I. Jordan
- Dustin Tran
- Geoffrey E. Hinton
In Statistical Learning Theory
- Francis R. Bach
- Michael I. Jordan
- Bernhard Schölkopf
- Kenji Fukumizu
- Alex Smola
- Arthur Gretton
- Per-Gunnar Martinsson
- Bharath K. Sriperumbudur
Citations in the corpus
- Fastfood: Approximate Kernel Expansions in Loglinear Time (2014, background): “See e.g. (Williams, 1998; Neal, 1994; MacKay, 2003) for details.”
- Being Bayesian, Even Just a Bit, Fixes Overconfidence in ReLU Networks (2020, background): “two-layer networks with infinitely wide hidden layers, are equivalent to Gaussian processes (Neal, 1996).”
- Reliable Uncertainty Estimates in Deep Neural Networks using Noise Contrastive Priors (2018, background): “Namely, GPs can be understood as the limit of infinitely wide networks (Neal, 1994; Matthews et al., 2018; Lee et al., 2018).”
- Enhanced Convolutional Neural Tangent Kernels (2019, background): “The correspondence between infinite neural networks and kernel machines was first noted by Neal (1996).”
- The Supervised Learning No-Free-Lunch Theorems (2002, background): “Often, for lack of an alternative, they do this without taking into account the ultimate effect on the direct object of interest, the input-output functions parametrized by those weights [10,6].”
- Unifying Divergence Minimization and Statistical Inference Via Convex Duality (2006, unknown): “– For ψ(t) = yψx(x) and y ∈ {±1}, we obtain binary Gaussian Process classification [15].”
- How to Scale Up Kernel Methods to Be As Good As Deep Neural Nets (2014, background): “Moreover, the connection between kernel methods and (infinite) neural networks has also been long noted [35, 51, 11].”
- Semiparametric latent factor models (2005, background): “The first path, due to [10], involved the observation that in a particular limit the probability associated with (a Bayesian interpretation of) a neural network approaches a Gaussian process.”
- Randomized numerical linear algebra: Foundations and algorithms (2020, background): “Neal (1996) proposed the idea of using empirical approximation for kernels arising in Gaussian process regression.”