Category: AI Terms
- QR Decomposition (Orthogonal-Triangular Matrix Factorization)
- LU Decomposition (Factors a matrix into two triangular matrices)
- ICA (Blind source separation via statistical independence)
- INT8 (Quantization of neural network weights and activations)
- PCA (Dimensionality reduction with variance preservation)
- SVD (Matrix factorization into three components)
- DLSS (Smart scaling and smoothing with a neural network)
- TOPS (Measuring peak performance of AI accelerators)
- BF16 (Floating-point number format for machine learning)
- FSR4 (Next-generation neural network scaling)
- Eigenvector (A vector that preserves direction under transformation)
- Eigenvalue (Scaling your own vector without rotation)
- Kronecker Product (Building a block matrix through scaling)
- Hadamard Product (Element-wise multiplication of matrices of the same size)
- GEMM (Matrix multiplication using the row times column method)
- Inner Product (Weighted sum of pairwise coordinate products)
- Dot Product (Summing products of two vectors)
- Transpose (Matrix row and column transposition)
- Shape (Returns the dimensions of a tensor)
- Rank (Number of dimensions (Axes) of a Tensor)
- Tensor (Multidimensional container for numerical data)
- Matrix (Storing data in tabular form)
- Vector (Ordered storage of numbers in continuous memory)
- Scalar (Converting a multidimensional tensor into a single number)