#Description A simple implementation of q-learning.
This is the scenario my q-learning implementation uses:
Where the blue line is a cliff that if the agent falls then the current episode ends a new one starts. S is the starting point and G is where it should go.
#Compile and run
To compile simply run:
gcc -o main main.c -Wextra
And then:
./plot.sh
#Sample result Result of 5 different runs. Each color represet a different execution.
Frequency of 5 different runs (not the ones above). By frequency I mean how many movements the agent needs to reach the objective.