Experimental features in the works at Steam can be tried by users thanks to a new feature called Steam Labs.
That's the idea behind Steam Labs at least. A discussion group for the new Interactive Recommender is available here. Just head over to Steam Labs, sign in to your account, and try one of three exciting new experiments. Micro trailers are pretty straightforward: Steam presents you with selections of six-second trailers organised by genre, curator selections, or other categories.
The automated show is assembled from similarly brief clips of games, but with multiple micro trailers for each game assembled in a quad display set to music. It's meant for you to take in and check out a few hundred games at a time, or leave on as background noise.
Automated Show - A half-hour video featuring the latest Steam launches. It's a "neural-network model that is trained to recommend games based on a user's playtime history, along with other salient data" based on "many millions of Steam users and many billions of play sessions".
"We view this new Interactive Recommender as one discovery element among many, and look forward to introducing more ways to connect customers with interesting content and developers", Valve wrote.
Interestingly, Valve doesn't give the model explicit information about games, but instead lets the machine learn about them during the initial training process. With Nvidia's Turing GPUs featuring Tensor Cores, itself hardware for the DLSS machine learning feature to save on performance through AI-powered image reconstruction, what do you think of the increased use of machine learning in games? But in a bit of behind-the-scenes information, Valve also explained that it doesn't feed any data about the game into the system except for the release date. Firefox has its Test Pilot program where you can try various in-development add-ons, and now Steam is getting something similar.
They conclude by saying that while this recommender is intended for consumers, they also wish to aid developers, "Developers can see how many page visits the recommender is generating directly in the existing "Traffic Breakdown" page for each game, though note that this experiment might not generate much traffic relative to the rest of Steam". Meanwhile, we'll be watching to see where this all goes.