Reinforcement Learning and surrogate reward functions based on graph Laplacians
MetadataShow full item record
Reinforcement learning is an upcoming area in machine learning with many applications. This thesis covers the basics of reinforcement learning: reward functions, value and policy iterations, and their algorithms. A value iteration algorithm for the game tic-tac-toe is given along with the results of a policy learning from itself. When the reward function is not straightforward to define, a surrogate reward function might be helpful. A surrogate reward function is defined by using the Fiedler vector of the Laplacian of the graph defined by the game. Laplacians based on weighted graphs in four different ways are defined and used to make different surrogate reward functions for a walking game. Finally, the surrogate reward functions are used in a value iterations algorithm and compared to the exact value function of the walking game.
Showing items related by title, author, creator and subject.
Rewards at Work: The Relationship between Rewards and Work Outcomes, and the Importance attached to Rewards Sares, S.M. (2020)The types of rewards one receives from work can have a huge impact on how employees see the organization they are working in. Some research has indicated that additionally the importance one places on these rewards might ...
Dopamine/Serotonin interaction in reward-related behavior: Focus on the Iowa Gambling Task Klanker, M. (2010)The Iowa Gambling Task (IGT) is a tool to study decision making behavior. Subjects choose cards from four decks; choosing a card will result in gain or loss of money and subjects are instructed to gain as much money as ...
The role of the orbitofrontal cortex in reward related behavior in addiction and alcoholism in particular Steketee, R.M.E. (2009)Addiction can be conceptualized as a learning disorder as addicts are unable to regulate behavior associated with drug reward. Reward related learning has traditionally been associated with dopamine transmission in subcortical ...