Skip to content

NathanR100/Wordle_Bot

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 

Repository files navigation

Abstract

Winning at Wordle is a complex task involving picking the correct letters to a 5 letter word from a wordbank and modifying guesses based on feedback (which letters are correct and which are incorrect). Our model attempts to solve this problem through reinforcement machine learning and a Deep Q-Learning algorithm. Our data included a random set of 5 letter words from all of the 5 letter words in the dictionary and the final model had a success rate of over 10 percent. This is more successful than most humans who play the game, proving the model's effectiveness. Future plans include increasing the computational capacity of our model through parallel processing as well as using Resnet to increase the depth of regularization.

Introduction

The New York Time's 'Wordle' is a popular word puzzle game where players attempt to guess a secret five-letter word by making guesses and receiving feedback on the correctness of each letter. Players have six attempts to correctly guess the word. Players receive feedback of green if the letter was correctly guessed in the correct position, yellow if the letter was correct however in the wrong position, and gray if the letter is not in the word. It is important to note that each guess must be a valid English word. For the New York Times 'Wordle,' the word of the day is the same for everyone. This makes for a widely enjoyed game that challenges players' linguistic and deductive skills. The goal in our particular project is to use reinforcement learning to correctly guess the 'Wordle' in under 6 attempts.

Creating an AI solver not only provides a tool for individuals to improve their gameplay but also serves as a fascinating application of reinforcement learning in solving word puzzles. Players, especially those passionate about word games, can benefit from an AI 'Wordle' solver by enhancing their strategic thinking and word-guessing abilities. It can be an educational tool for language learners and an entertaining resource for word game enthusiasts. 'Wordle' players looking to improve their skills and language learners seeking a fun and educational tool are both beneficiaries of this tool. This is also something that board puzzle enthusiasts interested in AI applications would be excited about.

The project's impact lies in creating a sophisticated AI system capable of solving 'Wordle' puzzles, showcasing the potential of reinforcement learning in puzzle solving and language-related tasks. The impact includes improved gaming experiences, educational benefits, and broader applications in natural language processing.

Related Work

Many have attempted to solve Wordle through algorithms, for example "Wordle is NP-Hard '' [3]. According to the paper "Wordle is NP-Hard" by Lokshtanov et al.[3], the challenge in Wordle can be described as an optimization problem where the objective is to minimize the number of guesses needed to deduce a hidden word. The NP-Hardness arises because, to optimally solve Wordle, one would need to analyze all possible combinations of guesses and feedback to determine the next best guess, akin to solving very complex decision trees. This exhaustive search escalates exponentially with the size of the vocabulary, making the problem computationally infeasible to solve optimally in polynomial time. Andrew Ho explains the computational complexity of Wordle,"Most of the state space is unreachable. One bound on the size of the state space is that there are 13k actions you could take at each turn. After each of these steps, the game can respond with one of 3^5 combinations of Noes, Maybes, and Yeses. So you could think of this as (13k * 3^5)^6 possible sequences, approx. 10^38"[1]. Other publications such as "Finding the optimal human strategy for 'Wordle' using maximum correct letter probabilities and reinforcement learning" [2] used reinforcement learning and Q-Learning to solve the problem. This paper also talks about "Calculating word log-likelihood based on green letter probability distribution" using the formula shown in figure 2.1 and "Calculating optimum Total Green Letter Probability (TGLP) for sequences of N words'' using the formula shown in 2.2.

Log- Likelihood

Total Green Letter Probability

While there are many methods that are popular in reinforcement learning to solve 'Wordle' such as policy gradient methods, model based reinforcement learning, and exploration and exploitation strategies, algorithms utilizing Deep Q-Learning or Advantage Actor Critic methods performed near optimal levels. Deep Q-Learning is an advanced reinforcement learning algorithm that integrates neural networks with Q-Learning, a model-free off-policy algorithm. In traditional Q-Learning, an agent learns to optimize the Q-value function, which estimates the value of taking a particular action in a given state. Deep Q-Learning enhances this approach by using a deep neural network to approximate the Q-value function, enabling it to tackle problems with high-dimensional state spaces. Advantage Actor-Critic methods are a type of policy gradient approach in reinforcement learning. They consist of two models: the Actor, which proposes actions given the state of the environment, and the Critic, which evaluates how good the action taken is, using a value function. The "advantage" part of A2C refers to the use of the advantage function, which measures the relative quality of an action compared to the average action in a given state, aiming to reduce the variance in the policy gradient estimation.

Data

The input being used to help develop a working 'Wordle' machine learning model has every 5 letter combination qualifying as a real word. The Kaggle dataset has almost 13,000 words listed in alphabetical order. However, of these 13,000 words only a small subset of over 2,000 words makes up the list of possible target words [1]. This subset of words removes many superfluous words such as "aahed" a very uncommon five letter word few are familiar with. However, though it may not be a solution, words such as "aahed" may be useful in discerning the actual target word. Due to the high computation complexity, our group starts with a random set of 10 words from the full dataset to train the model. These words were "spire", "crane", "slate", "champ", "ghost", "flint", "grace", "laugh", "sight", "bloom". Running the model with the original 13,000 words resulted in an extremely slow processing time and was inefficient in developing and testing our model. The use of 10 words were crucial in the development of our model as there were fast fun times, and we were able to see results, and fine tune accordingly.

Methods

The developed code used both reinforcement learning and Deep Q-Networks to train the agent to play 'Wordle'. Reinforcement learning was used because the team needed an agent to learn to interact with our environment that we developed to look like 'Wordle'. The Deep Q-Network was used to aid the construction of the reinforcement learning algorithm and make it accurate and efficient. By utilizing neural networks, our model approximates the Q-value function (figure 4.1.) reducing the computation complexity of our problem. Our algorithm used many packages including torch, deque, and generic packages such as pandas, numpy and random.

4.1. Q-value Function

As with all good reinforcement learning algorithms, the algorithm has an action, a state, a reward, a policy, and a value function (figure 4.1). The main reinforcement learning algorithm developed in our model was created from this generic reinforcement learning algorithm setup with the addition of Deep Q-learning. Our agent was a DQNAgent and the environment was the 'Wordle_Environment' class. The action was the guesses made right by the agent and the reward was calculated within the calculate_reward method of the 'Wordle' environment class. The policy was the Q-values and the value function was learned by the DQN model. In order to build the reinforcement algorithm, we used a few main classes. Our 'Wordle' environment class included various functions such as functions to help encode the feedback (listing green, yellow or black one hot encoded) as well as a reset to reset the game state and pick a new word from the list. The provide_feedback function gives feedback on the game and the update_state updates the game state. There are also functions to calculate the rewards as well as "step" to take a step in the game.

4.2. High Level Iteration

Unlike Q-Learning where the algorithm represents the quality of a certain decision, the deep Q-Learning network uses the value of the cumulative rewards for taking every action in the given state. It heavily rewards getting a green value resulting in a +3 for every green letter, mildly rewards yellow value with +1, and punishes black values with -1. Our forward method in the deep learning network passes the value forward through the neural network after we have defined the 'state_size' and 'action_size' that help define the game matrix and number of possible actions (respectively). The agent chooses the Q-value with the highest quality using a greedy algorithm. Overall, this algorithm is effective, gaining results slightly better than random guesses. However, this model has not been trained over millions of iterations as our group does not have access to resources to train the model at that level.

Thus, the neural network designed for our model was shallow and relatively simple to reduce computational complexity and allow experimentation. It begins with the input from the environment. The environment reports a sequence of 5 arrays with the following encoding - green: [1,0,0] yellow: [0,1,0] and black: [0,0,1] (figure 4.3.). This sequence is dependent on the feedback from the environment after a guess from the agent. The environment will return a tensor of five arrays in the position in which the environment returns the letter. These correspond to the colors above (green, yellow, or black). The input dimension for our neural network is the state size for each game. That is 5 letters, 3 colors, and 6 guesses, resulting in a total size of 90 for the input dimension.This feedback from the environment is then flattened into a list and fed into the first layer of the neural network.

4.3. Input

The structure of the neural network (figure 4.4.) is as follows: Input Layer - Feedback from Environment Layer 1 - ReLU activation function (figure 4.5.) with 128 nodes. Layer 2 - ReLU activation function (figure 4.5.) with 64 nodes. Output Layer - softmax activation function (figure 4.6.) giving probability that the agent should choose one of the items in the list of possible guesses.

4.4. Neural Network Structure

4.5. ReLU Function

4.6. SoftMax Function

The objective of the neural network is to maximize reward utilizing Mean Squared Error as our objective function. Mean Squared Error measures the squared difference between the estimated Q-values produced by the network and the target Q-values (computed from the rewards received and the next state's maximum predicted Q-values). The goal during training is to minimize this error, which aligns the predicted Q-values with the more accurate target Q-values derived from the Bellman equation(figure 4.6.)[4].

4.6. Bellman Equation

Exploration of decisions made by the agent are determined by the hyperparameter epsilon. Initially set to 1.0, this means the agent starts by exploring the environment fully, choosing actions randomly. This is crucial in the early phases of learning to ensure a diverse experience. Over time, epsilon is reduced by a small value after each episode, reducing the likelihood of random actions and increasing reliance on policy actions derived from the learned Q-values. Our epsilon decay value is set to 0.001. This decay continues until epsilon reaches a lower bound set to 0.1. This ensures that even as the agent becomes more confident in its learned policy, it retains some level of exploration to avoid local minima. The discount factor gamma, set to 0.5, moderates the importance of future rewards. A factor close to 1.0 places significant importance on future rewards, making the agent farsighted in its decision-making. This choice balances the agent's planning over a long horizon with short term reward, considering the long-term consequences of its actions versus instantaneous success. A balanced value was chosen as we are working with a reduced dataset size instead of the total 13,000 words.

To assist our DQNagent we have created the class ReplayBuffer. The ReplayBuffer is used to store the experiences of the agent, which includes states, actions, rewards, and next states. This buffer holds a fixed number of recent experiences. We had our buffer set to hold the same value of episodes at which the model would regularly report performance, which was usually a tenth of the total episodes simulated. Sampling from this buffer helps in breaking the correlation between consecutive learning updates, which can lead to more stable and effective learning. The use of a replay buffer also allows the agent to learn from individual experiences multiple times, which is more computationally efficient. Finally, the optimizer used is Adam. Known for combining the advantages of two other extensions of stochastic gradient descent, specifically Adaptive Gradient Algorithm (AdaGrad) and Root Mean Square Propagation (RMSProp). Adam is computationally efficient, requires little memory, and is well-suited for our problem.

Empirical Applications, Experiments, and Results

Our code was able to generate a success rate of around 10 percent over 500 episodes. While we were not able to run our model millions of times like many of the other models that we reviewed before building our model, a 10 percent success rate is indicative of successful learning in our model.

Cumulative Success Rate Over Episodes

5.2. Moves Per Episodes

![](data:image/png;base64,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

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors