Papers › Evaluating the Performance of TAAF for image classification models

Evaluating the Performance of TAAF for image classification models

13 Feb 2025- 2025 2archive 2025-07-28

Bryn T Chatfield

In this paper, we present the results of testing a custom activation function, The Analog Activation Function (TAAF), on both the MNIST and CIFAR-10 datasets. TAAF is a novel activation function designed to improve the performance of neural networks by leveraging a unique mathematical formulation. We evaluate TAAF in a convolutional neural network (CNN) architecture and compare its performance against standard activation functions on MNIST and against ELU on CIFAR-10. Our results demonstrate that TAAF achieves a test accuracy of 99.39% on the MNIST dataset and 79.37% on the CIFAR-10 dataset. On MNIST, TAAF achieves a slightly higher test accuracy of 99.39%, surpassing standard activation functions. On CIFAR-10, TAAF achieves a significantly higher test accuracy of 79.37% compared to ELU's 72.06% in the same architecture, suggesting improved generalization capabilities. This paper establishes a solid performance baseline for TAAF across different image classification tasks.

PaperPDFCode

Code

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Image Classificationimage-classification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Image Classification CIFAR-10 The Analog Activation Function Cross Entropy Loss 0.5551 #244 of 265 Archive leaderboard report
Image Classification CIFAR-10 The Analog Activation Function Parameters 545100 #244 of 265 Archive leaderboard report
Image Classification CIFAR-10 The Analog Activation Function Percentage correct 82.06 #244 of 265 Archive leaderboard report
Image Classification CIFAR-10 The Analog Activation Function Top 1 Accuracy 82.06 #244 of 265 Archive leaderboard report
Image Classification MNIST TAAF-CNN Accuracy 99.52% #31 of 81 Archive leaderboard report
Image Classification MNIST TAAF-CNN Cross Entropy Loss 0.0188 #31 of 81 Archive leaderboard report
Image Classification MNIST TAAF-CNN Epochs 35 #31 of 81 Archive leaderboard report
Image Classification MNIST TAAF-CNN Percentage error 0.48% #31 of 81 Archive leaderboard report
Image Classification MNIST TAAF-CNN Trainable Parameters 421642 #31 of 81 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Introduced by this paper: TAAF

ELUTAAF

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections