AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

ErfAct and Pserf: Non-monotonic Smooth Trainable Activation Functions

Koushik Biswas, Sandeep Kumar, Shilpak Banerjee, Ashish Kumar Pandey

AAAI 2022 · Architectures and efficiency › RNNs and normalization · approach: Theory of depth and width · 13 citations (OpenAlex)

Read the paper

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Replace ReLU with ErfActAlexNet · MNISTaccuracy: 99.09 $\pm$ 0.10 -> 99.51 $\pm$ 0.10Improved
Replace ReLU with PserfAlexNet · Fashion MNISTaccuracy: 93.22 $\pm$ 0.21 -> 93.82 $\pm$ 0.19Improved
Replace ReLU with ErfActWideResNet 28-10 · Tiny ImageNettop-1 accuracy: 61.61 $\pm$ 0.47 -> 64.20 $\pm$ 0.51Improved
Replace ReLU with PserfWideResNet 28-10 · Tiny ImageNettop-1 accuracy: 61.61 $\pm$ 0.47 -> 64.01 $\pm$ 0.49Improved
Replace ReLU with ErfActShuffleNet V2 2.0x · CIFAR100top-1 accuracy: 67.49 -> 73.17Improved
Replace ReLU with PserfShuffleNet V2 2.0x · CIFAR100top-1 accuracy: 67.49 -> 72.91Improved
Replace ReLU with ErfActShuffleNet V2 2.0x · CIFAR10top-1 accuracy: 91.63 -> 93.74Improved
Replace ReLU with ErfActU-Net · CityscapesmIOU: 69.32 $\pm$ 0.30 -> 71.29 $\pm$ 0.31Improved
Replace ReLU with PserfU-Net · CityscapesmIOU: 69.32 $\pm$ 0.30 -> 71.21 $\pm$ 0.34Improved
Replace ReLU with ErfActSSD300 · Pascal VOCmAP: 77.2 $\pm$ 0.14 -> 78.2 $\pm$ 0.12Improved
Replace ReLU with PserfSSD300 · Pascal VOCmAP: 77.2 $\pm$ 0.14 -> 78.2 $\pm$ 0.14Improved
Replace ReLU with ErfActTransformer · WMT-2014 newstest2014BLEU: 26.2 $\pm$ 0.15 -> 26.8 $\pm$ 0.11Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy