Looking for a solution to my below game problem. I believe it to require some sort of reinforcement learning, dynamic programming, or probabilistic programming solution, but am unsure... This is my original problem, and is part of an initiative to create "unique and challenging problem that you're able to conceptualize and then solve. 3 Judging criteria: uniqueness, complexity, and solution (no particular weighting and scoring may favor uniqueness/challenge over solution"
Inspirations: Conway's Game of Life, DeepMind's Starcraft Challenge, deep Q-learning, probabilistic programming
A bear is preparing for hibernation. A bear must reach life-strength 1000 in order to rest & survive the winter. A bear starts off at a health of 500. A bear explores an environment of magic berries. A bear makes a move (chosen randomly with no optional direction) and comes across a berry each time. There are 100 different types of berries that all appear across the wilderness equally and