1 code implementation • 18 Jun 2021 • Xinjie Lan, Kenneth Barner
However, it is intractable to accurately estimate the MI in DNNs, thus most previous works have to relax the MI bound, which in turn weakens the information theoretic explanation for generalization.
no code implementations • NeurIPS 2021 • Xinjie Lan, Kenneth Barner
The Information Bottleneck (IB) principle has recently attracted great attention to explaining Deep Neural Networks (DNNs), and the key is to accurately estimate the mutual information between a hidden layer and dataset.
no code implementations • 27 Oct 2020 • Xinjie Lan, Kenneth E. Barner
Based on the probabilistic explanations for MLPs, we improve the information-theoretic interpretability of MLPs in three aspects: (i) the random variable of f is discrete and the corresponding entropy is finite; (ii) the information bottleneck theory cannot correctly explain the information flow in MLPs if we take into account the back-propagation; and (iii) we propose novel information-theoretic explanations for the generalization of MLPs.
no code implementations • 16 Jun 2020 • Xinjie Lan, Xin Guo, Kenneth E. Barner
We study PAC-Bayesian generalization bounds for Multilayer Perceptrons (MLPs) with the cross entropy loss.
no code implementations • 22 Oct 2019 • Xinjie Lan, Kenneth E. Barner
Generalization is essential for deep learning.
no code implementations • 25 Sep 2019 • Xinjie Lan, Kenneth E. Barner
Based on the probabilistic representation, we demonstrate that the entire architecture of DNNs can be explained as a Bayesian hierarchical model.
no code implementations • 26 Aug 2019 • Xinjie Lan, Kenneth E. Barner
In this work, we introduce a novel probabilistic representation of deep learning, which provides an explicit explanation for the Deep Neural Networks (DNNs) in three aspects: (i) neurons define the energy of a Gibbs distribution; (ii) the hidden layers of DNNs formulate Gibbs distributions; and (iii) the whole architecture of DNNs can be interpreted as a Bayesian neural network.
no code implementations • 1 Jun 2019 • Xinjie Lan
In this paper, we propose a novel method for generating a synthetic dataset obeying Gaussian distribution.