Group Abstract Group Abstract

Message Boards Message Boards

6
|
741 Views
|
22 Replies
|
12 Total Likes
View groups...
Share
Share this post:
GROUPS:

[WSG76] Daily Study Group: Introduction to Game Theory

A Wolfram U Daily Study Group on game theory begins on Monday, March 30, 2026.

Join me and a group of fellow learners to learn about the basics of game theory. Our topics for the Study Group include:

  1. Simultaneous games
  2. Sequential games
  3. Repeated games

Make the right decision and learn the art of making decisions mathematically and rationally!

No prior Wolfram Language experience is required.

Please feel free to use this thread to collaborate and share ideas, materials and links to other resources with fellow learners!

Prisoner's Dilemma as a matrix game

March 30–April 3, 11am–noon CT (4–5pm GMT)

REGISTER HERE

Wolfram U Banner

Feel free to share this post on social media using the hashtags #WolframU and #DailyStudyGroup.

POSTED BY: Marc Vicuna
22 Replies

1) Is it possible to see the actual code that was used to create a tree diagram that was defined in the Function GameTheoryData[]? I realize you can display the tree structure by using (for example):

game = GameTheoryData["OptionalPrisonersDilemma"]["TreeGame"]
TreeGamePlot[game]

But how can I actually see the code behind the plot?

2) Are there any "Nature" tree games included in the Function GameTheoryData[] and if so, how would you identify them?

Thanks,
Mitch Sandlin

POSTED BY: Mitchell Sandlin

Hello Mitchell,

  1. More than the TreeGamePlot, the data structure hiding behind tree game is a Tree. If you want to have a deep control of a tree game, just call game["Tree"] to get its internal structure, and that should enable you to do a lot more. Unfortunately, a visual bug is on the current version (will be fixed soon!) that makes the name of the players hard to read, but you get the idea: enter image description here

  2. Yes, we do, namely the BeerQuiche game. You can see if any node has a probability as a tooltip, or you may notice the edges with probability have a different shade of gray and a different width than normal edges, although the case of 0.5 is probably the hardest to see. enter image description here

POSTED BY: Marc Vicuna
Posted 13 days ago

Using PlotLegends, I cannot seem to specify the text for the legend in the evolutionaryDynamicalSystemPlot function for the HawkDove Game in the last lesson.

Does the Legend need to be included in the formalization of the general function evolutionaryDynamicalSystemPlot before it can be formatted in the plot of a specific evolutionary game (example)?

If so, what is the appropriate Wolfram Language for that?

Thank you,

Steve Anderson

POSTED BY: Steve Anderson

Hello Steve,

All that is needed is to change the PlotLegends in the function itself yes, as seen here. The evolutionaryDynamicalSystemPlot function is relatively final, but if you need you could always take the series it inputs into a ListPlot and use your preferred visualization.

Best,

Marc Vicuna

POSTED BY: Marc Vicuna

I'm happy to announce that the new course has been released and is freely available in the Wolfram U catalog: Introduction to Game Theory

Thanks to everyone for their contributions during this development cycle. Sign in using your Wolfram ID (or create one to get started), and receive access to all the video lessons, quizzes, a sample exam, the final exam and certifications. The interactive course includes a progress tracker where you can monitor your progress toward a completion certificate. Take the final exam to earn a Level 1 certification for proficiency in game theory. Enjoy!

POSTED BY: Jamie Peterson
Anonymous User
Anonymous User
Posted 20 days ago

Hello. After the last class on game theory, all the participants have received an email for the recording. That email said "We will share a link to the Introduction to Game Theory framework, including access to the quizzes and final exam, in an email when the new interactive course is released."

I believe that we have not received this link yet. Will we receive the link soon? Also, if we want to solve the quizzes and pass the final exam, will we have enough time to do them?

Thank you.

POSTED BY: Anonymous User

The Wolfram U framework for this course will soon be available! Thank you for your patience.

POSTED BY: Marc Vicuna
Posted 21 days ago

Good day! I understand that the framework for the course will be available soon, but I'd like to ask if this will be before 14 April 2026. This is because my Wolfram Mathematica student trial will be expiring by then. Thank you so much! :)

POSTED BY: E G

I did not receive the email yesterday or today with the session recording. Are they available somewhere?

POSTED BY: Mark McBride

Hi, Mark! I forwarded the recordings to your email address, along with the additional information we shared during those sessions. You can also access them by clicking this link, which takes you to the series' landing page. Thank you for your participation in this Study Group!

POSTED BY: Roberto Martinez

Hi;

How would you set-up a matrix for the Rock,Paper,Scissors game?

Thanks,

Mitch Sandlin

POSTED BY: Mitchell Sandlin
POSTED BY: Marc Vicuna

Hi Marc;

This matrix works great; however, it looks like a vector if you view it in //MatrixForm, which was unexpected. Additionally, if you enclose each number of the matrix in another set of {} and view it using //MatrixForm, it looks the same. The matrix for the game works fine, I just didn't expect it to look like a vector and cannot understand why it looks like a vector.

Thanks,

Mitch Sandlin

POSTED BY: Mitchell Sandlin

I missed the study group session 4 due to I had another duty am I please still qualifying for the certificate ?

Hello Georgios Tsolakis,

As mentioned in the study group, the certification is based on passing the quizzes and final exam. Thus, when the course is available on Wolfram U, you'll be able to get your certification.

Best,

Marc Vicuna

POSTED BY: Marc Vicuna

Hi;

Why am I not getting the actions listed in my MatrixGamePlot function - see attached notebook?

Thanks,

Mitch Sandlin

Attachments:
POSTED BY: Mitchell Sandlin

I can see the actions with game["Dataset"]:

enter image description here

POSTED BY: Anton Antonov

Using "Dataset" displays the actions, but it is missing the players legend. What I am looking for is one graphic that shows all necessary information. If you are using the graphic to determine what action to take, you need the: players, actions, and payoffs to make the decision and shouldn't need to mentally combine several graphics to get the information.

Thanks,

Mitch Sandlin

POSTED BY: Mitchell Sandlin

Hi Mitchell,

As explained in the class, this was a design decision, as both representations emphasize different aspects of matrix games. However, to alleviate this issue, tooltips reveal both players and actions on both representations.
Worst case, if you need matrix games as a PNG, you could always legend the dataset fairly quickly, keeping in mind it would cease to be a Dataset.

Best,
Marc Vicuna

POSTED BY: Marc Vicuna

Question on exercise 2 - cheating students.
Proposed solution has these payoffs:

payoffs = {{{{0, 0, 0}, {0, 0, 1}}, {{0, 1, 0}, {0, -1, -1}}}, {{{1, 
      0, 0}, {-1, 0, -1}}, {{-1, -1, 0}, {-1, -1, -1}}}};

However, since this is a zero-sum, they need to be normalized. Is there any benefit to doing so vs defining payoffs as zero-sum integers like below?

payoffs = {{{{0, 0, 0}, {-1, -1, 2}}, {{-1, 
      2, -1}, {2, -1, -1}}}, {{{2, -1, -1}, {-1, 2, -1}}, {{-1, -1, 
      2}, {0, 0, 0}}}};
POSTED BY: Michael Elizarov

Hi Michael!

This was answered in class. Overall, zero-sum games have a few simple theorems, like the Minimax Theorem, the Equivalence of Nash Equilibria and Minimax Strategies, the Interchangeability of Equilibria, the Linear Programming Duality Formulation, the Saddle Point Characterization, the Zero Duality Gap (Strong Duality Interpretation), etc. that make them very easy to analyze mathematically. However, another thing to keep in mind is that some games can be transformed into zero-sum games as long as the ordering of payoffs is maintained. As long as the ordering is the same, the number of equilibria is also the same.

Best,

Marc Vicuna

POSTED BY: Marc Vicuna

In this delightful course, Marc Vicuna has skillfully blended a discussion of the computational tools for game theory in the Wolfram Language with an examination of the fundamental ideas of modern game theory.

I strongly recommend this short, one-week Study Group to everyone!

POSTED BY: Devendra Kapadia
Reply to this discussion
Community posts can be styled and formatted using the Markdown syntax.
Reply Preview
Attachments
Remove
or Discard