Our lives are shaped by all our past, current and future choices, but how do we make them? How can we determine which products to choose when online shopping? How do we identify the imposters in a game of Among Us? How do we develop theories that allow scientific advancement? To answer these questions, we rely on Bayesian inference – a fancy name for a rather straightforward process of updating one’s beliefs based on observed data in an attempt to understand the underlying causes of certain phenomena while taking into account uncertainties.
At the time of writing, I was in the middle of a homewards trip to Vietnam with my girlfriend. Having left our home country for over 3 years, we had always looked forward to buying lots of stuffs in Vietnam (because of nostalgia, and also because they’re much cheaper). With the rise of online shopping with home delivery, it is easier than ever to access a gigantic number of products, and the question of “How to buy” became “What to buy”.
Prior to our trip, my girlfriend had done a lot of research on many products, comparing similar options, reading and watching reviews, judging the pros and cons, etc. We never tried these products ourselves, so obviously the best bet was to collect as much information regarding these products as possible to make an informed choice.
Let’s pause for a moment and return to 2020 – you *probably* already know what I’m about to talk based on the title.
Amid the pandemic, Among Us saw a spike in popularity, and if you’re in your early 20s like me, *chances are* you had tried, or at least heard about this game. And if you have not, here’s the idea:
Among Us is an online multiplayer social deduction game, where you and your friends will take on the roles of cute little astronauts on a mission, and try to complete a certain number of tasks. However, there’s a catch: there will be one or more imposters among you who will try to sabotage the mission, and it’s up to the rest of the crew to collect evidence, identify the imposters and eject them into space before it’s too late.
Now, what does online shopping and Among Us have to do with one another, and what’s the deal with the Secrets of the Universe?
As curious beings, we humans have been observing nature for millennia, asking “Why” and making educated predictions. This process forms the basis of scientific research – we collect data, try to make sense of them and make predictions. Sometimes these predictions work, and sometimes they don’t, and in such case, we will need to update our beliefs in the form of new theories. For every so-called “Theory of Everything” developed, we hope to move one step further to find the truth of the universe.
Among these seemingly unrelated examples, there is an overarching theme of updating one’s belief based on evidence. This forms the heart of Bayesian inference. In Bayesian inference, the underlying cause of a certain phenomena is treated as a random variable, meaning it is not possible to identify a single absolute truth. It’s just that some cases are more likely than others in explaining the observed evidence.
The widely accepted product you found online may not suit your needs. The highly suspicious-looking guy you want to eject since the game started may be innocent. The so-called “Theory of Everything” may fail to explain some extreme cases. Perhaps, the universe may never fully unveil its secrets…
Dan Tran
Queensland University of Technology
