4 ms·

This comparison gets me the most. People bringing up human error rate vs machine error rate. They’re not comparable at all. And it will be evident when you have

by lvl100 5y ago

This comparison gets me the most. People bringing up human error rate vs machine error rate. They’re not comparable at all. And it will be evident when you have majority autonomous cars on the road. Human errors are more or less RANDOM. Machine errors are NOT.

machinelearning 5y ago
1. I’d challenge the premise that human errors are random. There are a ton of patterns that cause accidents including intoxication, low visibility conditions and tiredness. I haven’t done a statistical analysis but I’d hazard a guess that only a minority of human accidents are truly random. 2. Why does the randomness matter if the error rate is lower? Certainly if the errors are predictable, they can be discovered and fixed or avoided?
grlass 5y ago
also, human failure modes are better understood, and can be better anticipated by e.g. you can still have a somewhat predictive mental model of how a swerving drunk drive might behave. We don't have a good frame of reference for how machines might behave with their failures, which means that accidents could be worse than they would be otherwise.
- machinelearning 5y ago
  The severity of the accident is an interesting point. Though it intuitively feels like there are ways for software to mitigate the severity of an accident when it realizes it is about to crash than a human who might be asleep, intoxicated or otherwise have a slower reaction time.
  - kevingadd 5y ago
    The machine's ability to recognize that it's about to crash may actually be one of the issues here, since often the self-driving/driving-assist car crashes are cases where the AI just completely misinterpreted the environment and made bad choices. A human driver is somewhat likely to eventually realize what situation they've gotten themselves into (oh no, i can't stop in time) because of the multiple different feedback loops and information sources they're working with combined with their experience as a driver. For example, a drunk or very tired driver is operating with impaired decision making and response time, but they may eventually notice and respond - while an AI misclassifying a fire truck as a stop sign may very well continue misclassifying it until impact. One way to mitigate this would be via sensor fusion - even if your vision or radar sensing fail, you can rely on data from other sensors to do things like apply emergency braking. Unfortunately at least one vendor has decided to ditch radar, lidar, etc and just go with vision!
mainframed 5y ago
What do you mean with that, and why does the error type/"randomness" matter? If I can reduce the error rate by 90%, but the remaining 10% are "random" (whatever that means), is that worse than not reducing the error rate?
bluGill 5y ago
The only part that is random is if it gets you this time. People follow too close almost constantly, if that would cause an crash every time only in the most rural areas would people be able to drive even one mile without a crash. Note that the cause is pervasive: following too close. Others are bring up tired, drunk, texting... All real problems, but following too close is universal to nearly all drivers.

by lvl100 5y ago

machinelearning 5y ago

1. I’d challenge the premise that human errors are random. There are a ton of patterns that cause accidents including intoxication, low visibility conditions and tiredness. I haven’t done a statistical analysis but I’d hazard a guess that only a minority of human accidents are truly random. 2. Why does the randomness matter if the error rate is lower? Certainly if the errors are predictable, they can be discovered and fixed or avoided?

grlass 5y ago

also, human failure modes are better understood, and can be better anticipated by e.g. you can still have a somewhat predictive mental model of how a swerving drunk drive might behave. We don't have a good frame of reference for how machines might behave with their failures, which means that accidents could be worse than they would be otherwise.

machinelearning 5y ago
The severity of the accident is an interesting point. Though it intuitively feels like there are ways for software to mitigate the severity of an accident when it realizes it is about to crash than a human who might be asleep, intoxicated or otherwise have a slower reaction time.
- kevingadd 5y ago
  The machine's ability to recognize that it's about to crash may actually be one of the issues here, since often the self-driving/driving-assist car crashes are cases where the AI just completely misinterpreted the environment and made bad choices. A human driver is somewhat likely to eventually realize what situation they've gotten themselves into (oh no, i can't stop in time) because of the multiple different feedback loops and information sources they're working with combined with their experience as a driver. For example, a drunk or very tired driver is operating with impaired decision making and response time, but they may eventually notice and respond - while an AI misclassifying a fire truck as a stop sign may very well continue misclassifying it until impact. One way to mitigate this would be via sensor fusion - even if your vision or radar sensing fail, you can rely on data from other sensors to do things like apply emergency braking. Unfortunately at least one vendor has decided to ditch radar, lidar, etc and just go with vision!

mainframed 5y ago

What do you mean with that, and why does the error type/"randomness" matter? If I can reduce the error rate by 90%, but the remaining 10% are "random" (whatever that means), is that worse than not reducing the error rate?

bluGill 5y ago

The only part that is random is if it gets you this time. People follow too close almost constantly, if that would cause an crash every time only in the most rural areas would people be able to drive even one mile without a crash. Note that the cause is pervasive: following too close. Others are bring up tired, drunk, texting... All real problems, but following too close is universal to nearly all drivers.