Downloads · 30 days
0
Noisy-Gaussian-NN
GenAIDevTOProd/Noisy-Gaussian-NN
Noisy-Gaussian-NN is a machine learning model from GenAIDevTOProd. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as mit.
This project explores how a simple 1-hidden-layer neural network handles increasing label noise when fitting a Gaussian curve. We test three noise levels (σ = 0.05, 0.1, 0.2) to see when the network smooths effectivel…
Access
Public
Updated Aug 10, 2025
Repo size
—
Likes
0
Public
Repo makeup
Click a slice to open those files.
.ipynb297 KB · 99%
At a glance
- License
- mit
- Access
- Public
- Created
- Aug 10, 2025
- Updated
- Aug 10, 2025
- SHA
- 27b70c16
Model card
From the Hugging Face model README
Noisy Gaussian NN – Robustness to Label Noise
Overview
This project explores how a simple 1-hidden-layer neural network handles increasing label noise when fitting a Gaussian curve.
We test three noise levels (σ = 0.05, 0.1, 0.2) to see when the network smooths effectively and when it starts to underfit.
Dataset
- Synthetic dataset: Gaussian curve (
y = exp(-x^2)) - Noise added directly to labels using
torch.normal - 200 evenly spaced
xpoints in [-2, 2]
Model
- Architecture: 1 hidden layer, 50 neurons,
ReLUactivation - Loss: MSELoss
- Optimizer: Adam (lr=0.01)
- Training: 2000 epochs
Results
- Low noise: NN fits curve smoothly.
- Medium noise: Slight underfitting.
- High noise: Curve shape lost, noise dominates.
Key Insight
More noise ≠ better regularization.
Too much noise can destroy the signal beyond recovery.
Files
GaussianApproximation.ipynb– Full experiment, plots, and analysisREADME.md– This file
License
MIT License – free to use, modify, and distribute with attribution.
- License
- mit
- Languages
- en
- Created
- Aug 10, 2025
- Updated
- Aug 10, 2025