Questions tagged [connect-four]

For questions that involve the connect four game, which is a two-player connection board game, in which the players choose a color and then take turns dropping colored discs into a seven-column, six-row vertically suspended grid. The questions need to be related to AI.

5 questions
2
votes
1 answer

How should I define the reward function in the case of Connect Four?

I'm using RL to train a Network on the game Connect4. It learns quickly that 4 connected pieces is good. It gets a reward of 1 for this. A zero is rewarded for all other moves. It takes quite a time until the AI tries to stop the opponent from…
2
votes
0 answers

Reduction of state space of the game Connect Four to apply RL algorithms SARSA and Q-Learning

I would like to implement the reinforcement learning algorithms SARSA and Q-Learning for the board game Connect Four. I am familiar with the algorithms and know about their limitations regarding large states spaces due to storing all of this…
2
votes
0 answers

Which heuristic function should I use for the ColorShapeLinks game?

For learning purposes, I am trying to implement the minimax algorithm for the ColorShapeLinks game, which is similar to connect 4, except the fact that it combines both shape and color as the winning conditions, with a shape having priority over…
1
vote
1 answer

Deep Reinforcement Learning - How to Avoid a Naive Solution?

I'm implementing a connect-4 agent: using DQN, training by playing vs previous versions of the network. However, many times the network learns that it's best to simply put 4 pieces in the same column, which is usually able to beat the previous (also…
Ben Lahav
  • 13
  • 3
1
vote
0 answers

Negamax: how should you avoid the horizon effect in the connect four game?

I'm trying to implement a quiescence search in the negamax algorithm, for a connect four game. The algorithm is as follow for a chess game: int Quiesce( int alpha, int beta ) { int stand_pat = Evaluate(); if( stand_pat >= beta ) …
Carmellose
  • 151
  • 7