I attracted the most talented people to my team

This is my first time at a game jam! I hope to have a lot of fun and good experience.

explore procedurally generated islands, listen to the birds, and add to your bug collection in this relaxing little exploration game.

I'll be continuing to update until the end of the jam, but if you want to relax after a long day of coding, feel free to play it here! (just don't forget to come back and vote later ;) )

happy jamming everyone!

Talking about soothing shuffling of cardboard in a dumpster bin :D

but the cornmaze is causing some lag

edit: Build11 fixed everything im good to go :smile:

Day 2 and we almost got the core mechanic done... There is still much to do :cold_sweat:

... and we are too! We are making great progress with the little story elements we have planned for our game.

Always happy when I get to replace my default debug sphere with something fun!

With two hours to spare, my Compo entry Carrot Collector is out now! Munch your way through 12 levels as you try to collect all the carrots with only a limited number of moves!

Carrot Collector is playable on Newgrounds and itch.io, as well as here on the LD site. Please give it a look! As always, I will rate and review the games of everyone who rates and reviews mine.

Good luck to everyone still working on their games! You got this!

After several years of not doing any game development, a friend and I have decided to spend our large amount of free time this weekend making a game again!

My goal for LD is usually to learn something new. I embark on overly ambitious projects, but usually something nice comes out (although not necessarily within the deadline).

This time, I wanted to do something with level generation and reinforcement learning. I did a lot of level generation in the past, mostly using hand-crafted algorithms.

My inspiration was this paper: https://www.youtube.com/watch?v=ml3Y1ljVSQ8 They use Q-learning to train level generation. Q-learning requires a reward function for every state and action, which is in their case based on the solvability of a level. One of their examples is Sokoban, which I think is an interesting game to generate levels for. I wanted to up the ante a little, and decided to try this for Boulderdash. My eventual goal is to be able to generate levels for a larger class of puzzle-action games, such as CellSpace games (this is a tool I created during LD56).

So I created a level solver for a very basic Boulderdash game, based on breadth-first search. It's basically a brute force search, searching the entire state space. It comes up with the shortest path to complete a level. However, it turns out to be very slow. Even after some optimisations, it would slow down to >20 sec and eventually run out of memory for levels larger than about 6x6 tiles. This is because of combinatorial explosion: every step increases the search space up to 4-fold, because there are 4 possible actions (up, down, left, right).

I did not want to add game-specific optimisations to prune the state space. Instead I decided to try and use reinforcement learning to create a more universal (aka "model free") solver for larger levels. Last night, while I was asleep, I had my PC generate a dataset of optimal paths through 2500 random 6x6 levels using the brute force solver, serving to bootstrap my reinforcement learning system. If I can get this to work for larger levels than 6x6, it can generate new training data through self-play and support progressively larger levels (see the figure below).

Today I trained a neural net (a 5-layer Deep Q Network, based on convolution, with 2 million parameters) with part of the dataset, using Tensorflow JS, which I used before for browser games. TFJS support for CUDA sucks though, so I actually train it in the browser because it's the fastest. After some 20mins of training, it manages to find paths 100% of the time for seen levels, and about 60% of unseen (but known solveable) levels. It's clearly overfitted (apparently it has memorised the optimal paths of some 500 levels), but already has usable performance on 6x6 levels. However, on larger levels, it still falls down, with 10% performance on 8x8 levels and near zero for 10x10 levels. However this is still a first attempt, and there are a lot of things to try out, like hyperparameter optimisation and dataset improvement, so I'll be spending the next 24 hours trying to get this to work, and hopefully come up with a game with interesting generated levels!

This is my dog Buffy. Thanks for attention