Week 18

08/03/2024

Work done

• Conference paper abstract finished
• Dissertation work

Topics for meeting

• First draft of implementation section

Meeting Summary

• What makes an RL problem
• Discussed using delayed rewards
• Potential to change reward from S(s) to S(s') - S(s)
• Should there be a -1 reward for each time step (there is this already with gap penalty

Things to do

• Write a terminology section and stay consistent in dissertation
• Potentially work on a flexibility of RL section to discuss if this is an RL problem
• First draft of implementation section done