The Adventure agent stands at the entrance of a mysterious and treacherous cave, ands faint precious...

Transcribed Image Text:

The Adventure agent stands at the entrance of a mysterious and treacherous cave, ands faint precious gems, and including diamonds and -hearted; it's filled with obstacles, of Treasure exploring potest way port. The aim os, traps, 374-8 marked LOC A1. This cave contains lot of rubies. However, the cave is not for and challenges that the bravest adventurers can conquer. The agent is to reach the "exit" in the shortest possible and tries to grab as much as diamonds and rubies while exploring. The agent is equipped with four distinct actions: MoveUp, MoveDown, MoveLeft, and MoveRight. Each action incurs a cost of -5 for the a cell containing gems, it Furthermore, when the agent Cautomatically collects them. Additionally the act of grabbing a diamond yields a reward of +100, Conversely, if the in Conversely, dala043 diminishes, resulting in a [2+5=7 Marks] ent reaches auch action incor distinct 4374 agent enters a cell with a spider web, its grabbing power 10reward of cell grabbing a ruby provides a reward of +50. of -25. 74-86242 - Power 8-2022dal -2022da04374-86242 B 08-2022da04376 10/0 E 3 4 5 9/10/08-2022da EXIT 08-2022da0437 a. Construct partially filled Q-Table, Reward table and transition table. b. Apply the reinforcement learning with initial Q-Table initialized to value = 10. POST learning rate = 0.7 and factor=0.5 for the sequence of action listed below. It is mandatory to show the updated Q-Table at the end of every iteration. ng with initial and transition 08-2022da04 It is mandato and discount Perform MoveRight → MoveRight 04374-86242-le at the end of of action is da 10/08-2022 The Adventure agent stands at the entrance of a mysterious and treacherous cave, ands faint precious gems, and including diamonds and -hearted; it's filled with obstacles, of Treasure exploring potest way port. The aim oles, traps, 374-8 marked LOC A1. This cave contains lot of rubies. However, the cave is not for and challenges that the bravest adventurers can conquer. The agent is to reach the "exit" in the shortest possible and tries to grab as much as diamonds and rubies while exploring. The agent is equipped with four distinct actions: MoveUp, Move Down, MoveLeft, and MoveRight. Each action incurs a cost of -5 for the a cell containing gems, it Furthermore, when the agent Cautomatically collects them. Additionally the act of grabbing a diamond yields a reward of +100, Conversely, if the in Conversely, sala043 diminishes, resulting in a [2+5=7 Marks] ent reaches auch action indistinct 4374 agent enters a cell with a spider web, its grabbing power 10reward of 2 cell grabbing a ruby provides a reward of +50. of -25. 74-86242 - Power 8-2022dal -2022da04374-86242 B 08-2022da04376 10/0 E 3 4 5 9/10/08-2022da EXIT 08-2022da0437 a. Construct partially filled Q-Table, Reward table and transition table. b. Apply the reinforcement learning with initial Q-Table initialized to value = 10. POST learning rate = 0.7 and factor=0.5 for the sequence of action listed below. It is mandatory to show the updated Q-Table at the end of every iteration. ng with initial and transition 08-2022da04 It is mandato 2nd discount Perform MoveRight → MoveRight 04374-86242-le at the end of of action is da 10/08-2022 The Adventure agent stands at the entrance of a mysterious and treacherous cave, ands faint precious gems, and including diamonds and -hearted; it's filled with obstacles, of Treasure exploring potest way port. The aim os, traps, 374-8 marked LOC A1. This cave contains lot of rubies. However, the cave is not for and challenges that the bravest adventurers can conquer. The agent is to reach the "exit" in the shortest possible and tries to grab as much as diamonds and rubies while exploring. The agent is equipped with four distinct actions: MoveUp, MoveDown, MoveLeft, and MoveRight. Each action incurs a cost of -5 for the a cell containing gems, it Furthermore, when the agent Cautomatically collects them. Additionally the act of grabbing a diamond yields a reward of +100, Conversely, if the in Conversely, dala043 diminishes, resulting in a [2+5=7 Marks] ent reaches auch action incor distinct 4374 agent enters a cell with a spider web, its grabbing power 10reward of cell grabbing a ruby provides a reward of +50. of -25. 74-86242 - Power 8-2022dal -2022da04374-86242 B 08-2022da04376 10/0 E 3 4 5 9/10/08-2022da EXIT 08-2022da0437 a. Construct partially filled Q-Table, Reward table and transition table. b. Apply the reinforcement learning with initial Q-Table initialized to value = 10. POST learning rate = 0.7 and factor=0.5 for the sequence of action listed below. It is mandatory to show the updated Q-Table at the end of every iteration. ng with initial and transition 08-2022da04 It is mandato and discount Perform MoveRight → MoveRight 04374-86242-le at the end of of action is da 10/08-2022 The Adventure agent stands at the entrance of a mysterious and treacherous cave, ands faint precious gems, and including diamonds and -hearted; it's filled with obstacles, of Treasure exploring potest way port. The aim oles, traps, 374-8 marked LOC A1. This cave contains lot of rubies. However, the cave is not for and challenges that the bravest adventurers can conquer. The agent is to reach the "exit" in the shortest possible and tries to grab as much as diamonds and rubies while exploring. The agent is equipped with four distinct actions: MoveUp, Move Down, MoveLeft, and MoveRight. Each action incurs a cost of -5 for the a cell containing gems, it Furthermore, when the agent Cautomatically collects them. Additionally the act of grabbing a diamond yields a reward of +100, Conversely, if the in Conversely, sala043 diminishes, resulting in a [2+5=7 Marks] ent reaches auch action indistinct 4374 agent enters a cell with a spider web, its grabbing power 10reward of 2 cell grabbing a ruby provides a reward of +50. of -25. 74-86242 - Power 8-2022dal -2022da04374-86242 B 08-2022da04376 10/0 E 3 4 5 9/10/08-2022da EXIT 08-2022da0437 a. Construct partially filled Q-Table, Reward table and transition table. b. Apply the reinforcement learning with initial Q-Table initialized to value = 10. POST learning rate = 0.7 and factor=0.5 for the sequence of action listed below. It is mandatory to show the updated Q-Table at the end of every iteration. ng with initial and transition 08-2022da04 It is mandato 2nd discount Perform MoveRight → MoveRight 04374-86242-le at the end of of action is da 10/08-2022

Answer rating: 100% (QA)

Solutions Step 1 Explanation Part 1 Constructing Partially Filled QTable Reward Table and Transition Table To construct the QTable Reward Table and Transition Table we need to represent the cave as a ... View the full answer