Overlapping subproblems

Ever calculated a huge Fibonacci sequence by hand?

Imagine you're trying to figure out the 10th number in the Fibonacci sequence, but you keep recalculating the 8th and 9th numbers every time.

You'd waste a lot of time repeating calculations instead of remembering past results.

Example

To find F(10), you'd repeatedly calculate F(8) and F(9), even though you've already done it for F(8) to get F(9).

Remember this

Dynamic programming remembers past results to avoid redundant calculations.

Text adapted from Wikipedia, licensed under CC BY-SA 4.0.

Related concepts

Kolmogorov complexity

Kolmogorov complexity is uncomputable

Matrix multiplication algorithm

Ever wondered how computers speed up multiplying huge numbers?

Halting problem

Alan Turing proved the halting problem is undecidable

Greedy vs dynamic programming: greedy makes locally optimal choices, DP considers all subproblems

Greedy: locally optimal choices; DP: considers all subproblems

the Y combinator does: enables recursion in languages without named functions

Y Combinator launched over 5,000 companies

Computational complexity of matrix multiplication

O(n³) naive matrix multiplication

Swipe through 100 ML concepts daily