Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Smooth Min-Max Monotonic Networks

About

Monotonicity constraints are powerful regularizers in statistical modelling. They can support fairness in computer-aided decision making and increase plausibility in data-driven scientific models. The seminal min-max (MM) neural network architecture ensures monotonicity, but often gets stuck in undesired local optima during training because of partial derivatives of the MM nonlinearities being zero. We propose a simple modification of the MM network using strictly-increasing smooth minimum and maximum functions that alleviates this problem. The resulting smooth min-max (SMM) network module inherits the asymptotic approximation properties from the MM architecture. It can be used within larger deep learning systems trained end-to-end. The SMM module is conceptually simple and computationally less demanding than state-of-the-art neural networks for monotonic modelling. Our experiments show that this does not come with a loss in generalization performance compared to alternative neural and non-neural approaches.

Christian Igel• 2023

Related benchmarks

TaskDatasetResultRank
ClassificationCOMPAS (test)
Accuracy69.5
16
RegressionBlogFeedback SMM ICML-2024 regression benchmark suite (test)
RMSE0.154
10
ClassificationLoanDefaulter (test)
Test Accuracy65.47
10
ClassificationHeart Disease (test)
Test Accuracy91.3
7
RegressionAuto MPG SMM ICML-2024 regression benchmark suite (test)
MSE7.51
6
Disentangled Representation Learningsynthetic ellipses dataset
Reconstruction L2 Norm37
5
ClassificationChestXray (test)
Accuracy (Pretrained)67.9
4
Monotone multivariate function approximationSMM Monotone multivariate function approximation (Nin=2) ICML-2024 (test)
Median MSE0.00e+0
4
Monotone multivariate function approximationSMM (Nin=6) ICML-2024 (test)
Median MSE0.02
4
Monotone multivariate function approximationSMM Monotone multivariate function approximation (Nin=4) (test)
Median MSE0.01
4
Showing 10 of 10 rows

Other info

Follow for update