PC World - USA (2019-08)

(Antfer) #1
22 PCWorld AUGUST 2019

NEWS VALVE’S ‘STEAM LABS’ EXPERIMENTS


of New Releases, without even accounting for
popularity or some minimum quality
standard? That sounds like a nightmare. In any
case, I can’t imagine it’s more efficient than
simply going to Steam and clicking on the
games that sound even marginally interesting,
watching two seconds of trailer footage, and
closing out of the ones you aren’t into.

WELCOME TO THE MACHINE
The Interactive Recommender is Valve’s
biggest play, though, a replacement for the
current Discovery system that sounds—at least
on paper—significantly smarter and more
adaptable. “While existing store features like
tag-based searching can work well, we think
we can do better,” writes Valve.
And I certainly hope so. The current
Discovery system is better than nothing, but in
my experience recommendations tend to fall
into two categories, either serving up the
top-selling games on Steam or playing it really
safe. Valve remarks on the latter, noting that

“Just because you play a lot of
Beat Saber, doesn’t mean we
should only ever recommend
you VR rhythm games.” That is
sort-of how Discovery shakes
out though.
Valve’s new Interactive
Recommender instead draws
connections based on players,
not games.
“This model takes a
different approach. It disregards most of the
usual data about a game, like genre or price
point. Instead, it looks at what games you
play and what games other people play, then
makes informed suggestions based on the
decisions of other people playing games on
Steam. The idea is that if players with broadly
similar play habits to you also tend to play
another game you haven’t tried yet, then that
game is likely to be a good recommendation
for you.”
That’s the gist of it. There’s more detail in
Valve’s announcement, discussions of neural
networks and machine learning, but basically
the Interactive Recommender will suggest
games that aren’t necessarily similar to the
ones in your library, but are found in libraries
similar to yours.
You’re also able to filter its recommen-
dations, which is intriguing. You can sort by
tag of course, and by release date. However
you can also sort by what Valve’s calling
“Popularity.” Via Valve:

The Automated Show.
Free download pdf