Anomaly Detection Quotes

I think all scientists are like detectives. We are most happy when we find something that doesn't fit our expectations.

A lot of time mistakes are very interesting - you look for the behaviour that's not the one you expect.

A computer chatted to itself in alarm as it noticed an airlock open and close itself for no apparent reason. This was because Reason was in fact out to lunch.

We've taken disturbances and fluctuations and averaged them together to give us comfortable statistics. Our training has been to look for big numbers, important trends, major variances. Yet it is the slight variations - soft-spoken, even whispered at first - that we need to encourage.

If I see something that seems out of sync with what's already known, the first thing I do is try to find out what's wrong with the data. Once you've done that, and it still seems wrong, that's when things get interesting. It means you've found something new to understand.

Normal science, the activity in which most scientists inevitably spend most all their time, is predicated on the assumption that the scientific community knows what the world is like. Normal science often suppresses fundamental novelties because they are necessarily subversive of its basic commitments.

I think our science is a marvelous tool. Of course it has its politics, its failings, its mistakes, like all other human endeavors, but I think the methodology of our science - using it to postulate and to use the proper approach of creating falsifiable hypotheses and then testing them to see if they ring true, and then progressing by finding the anomalies to our theoretical structure, and then improving our theoretical structure to take care and test the anomalies - is the way to go, because that helps us discover what the universe.is all about and our relationship to it.

In other words, by finding the anomalous event, what you do is you get out ahead of activities.

But by providing the background picture - the universal situational awareness that we desire - by showing the anomalies, the Space-Based Radar will change the nature of how we do our analysis and our intelligence.

The only thing police patrol cops - in certain situations - are expert at is spotting anomalies. When you are a black person that is driving in a place that you stick out, that's all they're going to see.

Often a sign of expertise is noticing what doesn't happen.

You want to know what a robot's designed for. And if it's doing something outside the scope of what it's made to do, you should be very suspicious.

Our sense of safety depends on predictability, so anything living outside the usual rules we suspect to be an outlaw, a ghoul.

If the theory accurately predicts what they [scientists] see, it confirms that it's a good theory. If they see something that the theory didn't lead them to believe, that's what Thomas Kuhn calls an anomaly. The anomaly requires a revised theory - and you just keep going through the cycle, making a better theory.

Know what is behind you, and pay particular attention to anything out of place.

We have to do what I would call anomalies: we have to look for strange things that show up once in a while. They don't show up all the time. We have to be scanning the horizon, and doing that, once in a while something will show up that makes a lot of sense, and then you act on it.

We search for patterns, you see, only to find where the patterns break. And it’s there, in that fissure, that we pitch our tents and wait.

If you want to make better theory, you've got to use the best that's available and look through the lens of another discipline to see if you can uncover more anomalies. By looking at the phenomena of failure from the perspective of sales, marketing, finance, general management, and the equity markets, I was able to see things that Rebecca [Henderson] hadn't.

Comedians are innately programmed to pick up oddities like mispronounced words, upside-down books on a shelf, and generally undetectable mistakes in everyday life.

I buy about $1,500 worth of papers every month. Not that I trust them. I'm looking for the crack in the fabric.

If you are looking for a needle in a haystack, and somebody has already cataloged all the straw in the haystack, when you get to that needle you will recognize it's different than what was supposed to be there based on all that computerized haystack information that had been predetermined for you.