weight
A weight is a learned scalar or tensor parameter that modulates the contribution of an input or intermediate signal through a linear or affine transformation, determining how strongly it influences the model’s output.
In neural networks, weights connect units within and across layers and, along with biases, are updated during training by gradient-based optimization.
Step a toy network through a forward pass, a backward pass, and a weight update, and watch the connection weights thicken and fade as gradient descent adjusts them:
They’re typically initialized from specific distributions and may be regularized with techniques like weight decay. They can be shared across positions as in convolutions or attention projections.
In practice, weights also refers collectively to a trained model’s full set of learned parameters. That artifact is what gets released, downloaded, quantized, and fine-tuned, as in an open-weights model.
Related Resources
Tutorial
Python AI: How to Build a Neural Network & Make Predictions
In this step-by-step tutorial, you'll build a neural network from scratch as an introduction to the world of artificial intelligence (AI) in Python. You'll learn how to train your neural network and make accurate predictions based on a given dataset.
For additional information on related topics, take a look at the following resources:
- Linear Regression in Python (Tutorial)
- Building a Neural Network & Making Predictions With Python AI (Course)
- Stochastic Gradient Descent Algorithm With Python and NumPy (Tutorial)
- Python AI: How to Build a Neural Network & Make Predictions (Quiz)
- Starting With Linear Regression in Python (Course)
- Linear Regression in Python (Quiz)
Have a question about this? Mentor AI can show you examples, compare related terms, and point you to tutorials.
By Leodanis Pozo Ramos • Updated Nov. 18, 2026