Hello whoever reading this, if you're here to take notes from this video... here you go. Models: Something that represents something else (replica), models allow predictions to be made about the object/event being represented - how it will behave under different conditions. Eg: Climate models show us the current climate and allow us to predict what will happen next. Models: - Something made out of lego or modelling clay (physical model) - Scaled model of a planned new building (visual model) Scientific Models (Models that scientists use) - Complex mathematical expressions (mathematical model) - Things stored on computer: weather, data (computer model) - There are also physical models and computer models in science Good sides of models (pros) - Models can be used to predict hazards to protect our health. - Models calculated our exposure to risky things, also protects our health. Bad sides of models (cons) - Models can be dangerous if it is a bad model and it is used inappropriately - Models have their limits since models aren't a complete representation of what they're representing (may give out false information) Other facts - Animals are also models. - Models are very powerful tools
@riskbites
3 жыл бұрын
Thanks!
@lucapasvolsky1677
2 жыл бұрын
your my favorite youtuber keep doin what your doing
@johnlime1469
4 жыл бұрын
I hope the uploader responds to this, but I think a huge problem that I had with the field of deep reinforcement learning when I was studying that field is that they combine all of these deep neural networks to make a hierarchical structure that sort of are able to complete certain robotic control tasks, but we have no clue as to when those models fail, why they fail when they do, and have no clue what the neural networks were able to learn from them. There is no insight given from these generated neural networks. It's such a huge black box that you throw at a problem and hope that it works. I hope these neural network models will be used not as the model itself, but a statistical tool that can analyze how physical or biological systems operate, so that we can build a structural model that we can understand from.
Пікірлер: 8