Intelligent-Agent

EPFL IA Project

Reactive Agent

We are required to solve Pickup and Delivery Problem using reinforcement learning (Markov Decision Processes) to learn an optimal strategy

Deliberative Agent

Implement the state-based breadth first search and $A^{*}$ heuristic search algorithms.

Centralized Agent

Implement the Stochastic Local Search algorithm for the PDP.

Auctioning Agent

The competition rules are as fellows,

1. The auction house publishes the details of the task, i.e. the pickup city, the delivery city, and the weight of the task are revealed to the agents.

2. Every interested agents may then submit bids for the task. The bids are integer numbers, representing the payment that the agent requests for the delivery of the task.

3. The task is assigned to the agent with the lowest bid and that agent is paid according to his bid.

The task is design and implement the best strategy to beat others.

Our Team(G46) win 2rd in the tournament as it described below,

Here are the tournament results and the associated bonus points:

Groups who qualified to the 1st round (5pts):
G20, G25, G34, G48, G21, G28, G5, G26, G30, G11, G14, G44, G39, G36, G8, G6, G24, G27, G23, G15, G49, G46, G29, G17, G7

Groups who qualified to the 2nd round (+10pts):
G20, G21, G28, G05, G26, G14, G36, G08, G27, G15, G49, G46

Groups who qualified to the 3rd round (+20pts):
G21, G05, G14, G36, G27, G46

3rd group (+10pts): G21

2nd group (+20pts): G46

Winner (+30pts): G36

(If you spot any mistake, please let us know.)

Thank you for participating.