Fetching the paper…

Reading the bibliography…

2022

Expressing linear equality constraints in feedforward neural networks

Rangarajan, Anand, He, Pan, Lee, Jaemoon et al.

Original
PDF
Post on X

Understand

We seek to impose linear, equality constraints in feedforward neural networks.

As top layer predictors are usually nonlinear, this is a difficult task if we seek to deploy standard convex optimization methods and strong duality.
To overcome this, we introduce a new saddle-point Lagrangian with auxiliary predictor variables on which constraints are imposed.
Elimination of the auxiliary variables leads to a dual minimization problem on the Lagrange multipliers introduced to satisfy the linear constraints.

Expressing linear equality constraints in feedforward neural networks · Around

Built on

The relaxation method of finding the common point of convex sets and its application to the solution of problems in convex programming
Bregman, L. (1967) · 1967
Earlier work this paper cites.
Algebraic transformations of objective functions
Mjolsness, E. and Garrett, C. (1990) · 1990
Earlier work this paper cites.

Similar

Statistical physics algorithms that converge
Yuille, A. L. and Kosowsky, J. J. (1994) · 1994
Cited alongside, same era.
Convex Optimization
Boyd, S. P. and Vandenberghe, L. (2004) · 2009
Cited alongside, same era.
Mémoire sur l’intégration de quelques équations aux différences partielles
Legendre, A. M. (1787)
Cited in the paper.

Then

On conjugate convex functions
Fenchel, W. (2014) · 2014
Later among the works it cites.
Lagrange multipliers revisited
Slater, M. (2014) · 2014
Later among the works it cites.

Beyond the bibliography

alphaXiv searches the wider corpus for related work and actual follow-ups.

Open on alphaXiv

alphaXiv is searching for related work…