A Tensor is essentially a container for numbers organized in multiple dimensions, like a more advanced version of a spreadsheet. While a single number is just a value and a list is a row of numbers, a Tensor can organize data in grids, cubes, or even higher dimensions—imagine a photo as a grid of colored pixels, or a video as a stack of those grids over time. Tensors are the building blocks AI systems use to store and process all kinds of information, from text and images to sound and sensor data. Modern AI frameworks like TensorFlow and PyTorch are built around tensor operations, performing efficient mathematical computations on these multi-dimensional arrays.
Get the latest news, advances in research, policy work, and education program updates from HAI in your inbox weekly.
Sign Up For Latest News

With corporate AI labs turning inward, academia must carry forward the mantle of open science.
With corporate AI labs turning inward, academia must carry forward the mantle of open science.


This year’s report shows a maturing industry, significant private investment, and rising competition between China and the U.S.
This year’s report shows a maturing industry, significant private investment, and rising competition between China and the U.S.
