TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

The Automated Dungeon Master

89 pointsby cfmcdonaldover 5 years ago

11 comments

Animatsover 5 years ago
That went on for a long time before getting to the point.<p><i>&quot;Trained on gigabytes of input, GPT-2 is uncannily good at producing sensible text by simply predicting the next word that should follow a given list of input words.&quot;</i><p>There are now a few systems like that - autocomplete with a big text database. They have no underlying model of what they&#x27;re talking about. So, after about three sentences, you realize the output makes no sense. It&#x27;s just words strung together with probabilities that match known text.<p>This seems to be where machine learning gets stuck. There needs to be some underlying model of the subject matter to have a useful dialog. Outside of well defined subject areas (sports, weather, travel planning, shopping, etc.) that barely exists. Frustrating to see this half a century after Eliza.<p>Chatbots have the same problem. Either they force you onto a specific track, like a phone tree, or they just natter endlessly without going anywhere. I&#x27;ve been playing around with Rasa lately. This is a chatbot which uses Tensorflow to match user questions with canned answers. That&#x27;s about all they get from the machine learning part. Outside of that it&#x27;s a phone tree engine. There&#x27;s a file of &quot;smalltalk&quot; questions and answers, so it can natter better.<p><i>What do we want? Chatbots!</i><p><i>When do we want them? Sorry, I don&#x27;t understand the question.&quot;</i> - bag at chatbot conference.
评论 #22137464 未加载
评论 #22137605 未加载
评论 #22138728 未加载
评论 #22138619 未加载
评论 #22134421 未加载
cfmcdonaldover 5 years ago
Thanks for the feedback, everyone.<p>A few of you mentioned Dwarf Fortress. I am aware of it, and considered adding a section on it after Elder Scrolls, but this was already twice the length of one of my usual posts, and I wanted to &#x27;ship it.&#x27;<p>It was pointed out that I only presented two alternatives in my conclusion, but there&#x27;s a third of a super-dense procedurally generated world a la Dwarf Fortress. I agree this was an oversight on my part. This points in the direction of an interesting subtlety about whether a rigorous simulation is what you really want at all - the DM &#x27;simulates&#x27; a world in a way that is highly biased towards adventure, excitement and fun.<p>I may go back and revise to try to bring in some of these points at a later date.<p>One person mentioned the Mythic GM simulator, I&#x27;m aware of that also, but a) it&#x27;s a very niche product and b) it still requires <i>someone</i> to GM, that person may just also be the sole player. It does provide a biasing function so that you can by &#x27;surprised&#x27; by outcomes and guide your own adventure in unplanned directions. It actually is more in the realm of a DM aid, as some have mentioned this may be the most fruitful direction for AI, as an aid rather than a replacement for a DM.
jammygitover 5 years ago
As a dungeon master, I have to say I did not expect my job to be automated so quickly
评论 #22137465 未加载
qzncover 5 years ago
I agree with the conclusion that AI dungeon might be the first glimpse of new kind of roleplaying game.<p>One tiny point i miss in the article is a reference to Dwarf Fortress, probably the most advanced fantasy world simulation.
评论 #22136804 未加载
评论 #22139705 未加载
ogre_codesover 5 years ago
It is easy to see a procedurally created game replacing a mediocre GM.<p>But it&#x27;s extremely difficult to see any kind of procedurally created game being able to create the sort of experience you get with a really good GM.<p>It is the difference between writing a story together or playing along in someone else&#x27;s story.
评论 #22133502 未加载
评论 #22133880 未加载
评论 #22133790 未加载
评论 #22137697 未加载
Fjolsvithover 5 years ago
Probably the game that is closest to being full AI would be Dwarf Fortress. You can play it in Adventure mode and you can literally talk with NPC&#x27;s who all have backstories and lives.
评论 #22135691 未加载
评论 #22136906 未加载
Mountain_Skiesover 5 years ago
When I saw Telengard played on a friend&#x27;s Commodore 64, I thought for sure it was the end of Dungeons &amp; Dragons forever. Who would want to bother with getting a DM and buying lots of books when you could get the same thing from a computer whenever you wanted?<p>Glad I was wrong!
评论 #22139366 未加载
mlillieover 5 years ago
Not including the &quot;Mystic&quot; system seems like a huge oversight. It&#x27;s an analog game system that works using yes&#x2F;no questions, 2D10, and a probability table. It&#x27;s not too elegant, but it works wonders. It can be used to assist the DM, or as a full-on DM emulator.
评论 #22136362 未加载
echan00over 5 years ago
This is a great article. I&#x27;ve been considering where this will head with gpt-2 dungeon master showing how far AI has come.<p>While I agree with the article that so far everything digital has been a noble failure, I feel there is too much discussion between AI vs humans. I think the future is in AI-assisted human DMs. Software may not be able to &quot;entertain and dazzle&quot; with the creativity of humans but it can help coordinate many routine tasks in any dnd game allowing the human DMs to focus their time on the components that make the adventure fun.
评论 #22134243 未加载
donparkover 5 years ago
I think there are other forms of dungeon mastering that use little or no AI while delivering the same if not better experience.
jmiskovicover 5 years ago
Great article about a topic that has fascinated me for quite some time. Please excuse the wall of text.<p>Procedural generation can provide unlimited amounts of content seeded from finite effort. It&#x27;s still mostly hand-crafted, just at different abstraction level. Building blocks are constructed manually and algorithm that fits them together is also carefully designed and fine tuned. It is used often in indie games because of lower effort needed to give illusion of huge amount of content. Meanwhile AAA companies can afford armies of level designers to make a detailed words. I guess everyone agrees that human design beats procedural and current AI. It is more varied, imaginative and purposeful - the 3 ingredients necessary for immersion. The crowd-sourced approach has not yet been successfully applied, to best of my knowledge.<p>The work that GM is doing can be quite valuable, and yet it only reaches few players of that campaign and is remembered vaguely. What if we could capture and store GM responses to player prompts? What if other sessions could reuse these responses and build on them to create new responses? The basic use of recorded responses is to automate mundane actions and checks (fights, lockpicking...) so that GM can focus on creative stories and arcs.<p>I&#x27;ve thought about this for quite some time and came up with partial solution. The automatic response system could consist of rules that are hand-crafted by GM while the session is played. Each rule is response to some action and changes the world state. So, each rule has to store the name of action, portion of world state that GM considers critical for that action, and how the world state is affected by action.<p>For example, let&#x27;s attack an orc with a sword. OK, the GM decides that player&#x27;s strength, sword sharpness, orc&#x27;s agility and a dice role are important in this interaction, so GM marks them as such. Because D20 dice came out as 12, GM decides that attack was fairly successful, and changes orc&#x27;s state to severely injured. This interaction is saved and the gaming session moves on. Later on, the same interaction &#x27;attack with sword&#x27; is repeated by with different world state. If important parts of world state are the same&#x2F;similar, then same result can be applied. If GM decides that existing rule is not applicable, then differentiating world state is marked and different results are entered and stored under a new rule. Now system is has two rules for &#x27;attack with sword&#x27; and can automatically choose between them, based on which rule matches the world state the most. With time the system becomes rich enough to simulate this action under various conditions, while still being able to accommodate new situations with help from GM.<p>Such rule-based system can be visualized as bunch of simple state-charts that are built on the fly, and they interact together by modifying global state. I can imagine this simple mechanism would be good enough for combat mechanics, crafting system, simple NPC simulation (moods, dialogs, quests), inventory interaction, terrain manipulation... It should be enough to hand-craft a rich immersive world?<p>The hard part is coming with data structure that can capture rich world state in enough details without overwhelming the GM. The structure would have to model locations with objects, each with its attributes. All this would have to be relative to character who wants to preform the action. I&#x27;m considering graph databases, but don&#x27;t have enough experience with them to move onto implementation :(