Polynomial self-attention

A method for allocating a raw image to a respective class out of a plurality of classes, the method comprising:

• receiving a raw image;
• generating a plurality of raw matrices from the raw image, wherein each raw matrix of the plurality of raw matrices is formed by one spatial value and one characteristic value;
• and
• allocating the raw image the respective class out of the plurality of classes based on the plurality of raw matrices.

An apparatus for training an image classification model, the apparatus comprising one or more processors configured to:

• receive a training image;
• generate a plurality of training matrices from the training image, wherein each training matrix of the plurality of training matrices is formed of one spatial value and one characteristic value;
• and
• adapt the image classification model based on the plurality of training matrices.

A method for training an image classification model, the method comprising:

• receiving a training image;
• generating a plurality of training matrices from the training image, wherein each training matrix of the plurality of training matrices is formed of one spatial value and one characteristic value;
• and
• adapting the image classification model based on the plurality of training matrices.