Tile Coding Index

Problem

Compute the tile-coding index for a continuous state.

Each dimension maps to int((state[i] + offsets[i] * widths[i]) // widths[i]) mod n_tiles[i], then dimensions are combined with mixed-radix strides, last dimension fastest. Return the integer index.

Hint

Offsets shift the grid so different tilings cover different boundaries.

Test cases

case 1

in: [[0.5,0.5],[0.1,0.1],[0,0],[10,10]]

expected: 44

case 2

in: [[0.95,0.05],[0.1,0.1],[0,0],[10,10]]

expected: 90

case 3

in: [[0.5,0.5],[0.1,0.1],[0.5,0.5],[10,10]]

expected: 55

case 4

in: [[0.25],[0.5],[0],[4]]

expected: 0

Starter codePython

def tile_coding_index(state, widths, offsets, n_tiles):
    # Your code here
    pass

4 test cases · run your code against them in the editor