4 ms·
> i wonder how much money gets wasted on these failed explorations/projects. Failure is not a waste of money if lessons are learned. The value is in the lesson
by mojomark 6y ago
> i wonder how much money gets wasted on these failed explorations/projects.
Failure is not a waste of money if lessons are learned. The value is in the lesson. As anyone whosed developed a reinforcement learning application knows, there is a balance between EXPLORATION and EXPLOITATION. If you don't explore, your performance will never improve. This is true even if 99% of the explored states are failures.
I suppose if one doesn't want to invest in exploration, one could always join an Amish township. They are very content and apparently happy with their perpetually static state of technological development - and there's nothing wrong with that until someone with a missile comes in to take over your land amd all you have to fend them off with is a hay fork.
- b3kart 6y ago> if lessons are learned This is the key statement. Even in reinforcement learning exploring intelligently is an important problem. There’s no point exploring a new state if your world model is confident about what’s going to happen there: you won’t learn much.