paper-with-me

Papers

Neural network identifiability for a family of sigmoidal nonlinearities

2019-06-11 · Verner Vlačić, Helmut Bölcskei

This paper addresses the following question of neural network identifiability: Does the input-output map realized by a feed-forward neural network with respect to a given nonlinearity uniquely specify the network architecture, weights, and biases? Existing literature on the subject Sussman 1992, Albertini, Sontag et al. 1993, Fefferman 1994 suggests that the answer should be yes, up to certain symmetries induced by the nonlinearity, and provided the networks under consideration satisfy certain "genericity conditions". The results in Sussman 1992 and Albertini, Sontag et al. 1993 apply to networks with a single hidden layer and in Fefferman 1994 the networks need to be fully connected. In an effort to answer the identifiability question in greater generality, we derive necessary genericity conditions for the identifiability of neural networks of arbitrary depth and connectivity with an arbitrary nonlinearity. Moreover, we construct a family of nonlinearities for which these genericity conditions are minimal, i.e., both necessary and sufficient. This family is large enough to approximate many commonly encountered nonlinearities to within arbitrary precision in the uniform norm.

📄 PDF Abstract BibTeX arXiv:1906.06994

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Affine symmetries and neural network identifiability

2020-06-21 · Verner Vlačić, Helmut Bölcskei

We address the following question of neural network identifiability: Suppose we are given a function $f:\mathbb{R}^m\to\mathbb{R}^n$ and a nonlinearity $\rho$. Can we specify the architecture, weights, and biases of all …

Independent Innovation Analysis for Nonlinear Vector Autoregressive Process

2020-06-19 · Hiroshi Morioka, Hermanni Hälvä, Aapo Hyvärinen

The nonlinear vector autoregressive (NVAR) model provides an appealing framework to analyze multivariate time series obtained from a nonlinear dynamical system. However, the innovation (or error), which plays a key role …

Time SeriesTime Series Analysis

Beyond Gaussian Initializations: Signal Preserving Weight Initialization for Odd-Sigmoid Activations

2025-09-27 · Hyunwoo Lee, Hayoung Choi, Hyunju Kim arxiv

Activation functions critically influence trainability and expressivity, and recent work has therefore explored a broad range of nonlinearities. However, widely used Gaussian i.i.d. initializations are designed to preser…

Automated Design of Linear Bounding Functions for Sigmoidal Nonlinearities in Neural Networks

2024-06-14 · Matthias König, Xiyue Zhang, Holger H. Hoos, Marta Kwiatkowska 외

The ubiquity of deep learning algorithms in various applications has amplified the need for assuring their robustness against small input perturbations such as those occurring in adversarial attacks. Existing complete ve…

Local Dynamics in Trained Recurrent Neural Networks

2016-07-13

Learning a task induces connectivity changes in neural circuits, thereby changing their dynamics. To elucidate task related neural dynamics we study trained Recurrent Neural Networks. We develop a Mean Field Theory for R…