Finite element method

second-order methods (Newton's) converge faster but are expensive: O(n³) per step

Second-order methods converge faster due to quadratic convergence but are expensive due to O(n³) per iteration

Euler method

Euler method approximates ODE solution with y_{n+1} = y_n + h·f(y_n)

iterative methods (CG, GMRES) do: solve Ax=b without explicitly inverting A

CG, GMRES iteratively solve Ax=b without explicitly inverting A

quantization to INT8 doubles throughput

Quantization to INT8 doubles throughput because tensor cores process INT8 2x faster

approximation algorithms guarantee: solution within factor α of optimal

Approximation algorithms guarantee a solution within a factor α of the optimal solution

Ordinary least squares

OLS minimizes squared differences