- MOF derived supercapacitor electrode material
- Should I italicize the name of a paper when referring to it in an email?
- How can I create a chart like CandlestickChart, but for non-financial data with something other than dates as the abscissæ?
- WSTP examples hang after receiving input
- Numerical integration with a variable
- Why NSolve giving answer even though there isn't any
- Using ReadList with Throw
- How to hie “input” while using ButtonBar?
- how can i color in mathematica?
- Please identify this bird
- Please explain this notation MW around 9 MDa
- What is the significance of Nimrod to Dunkirk's original score?
- What are the film techniques for broken and protruding bones?
- Why did they kick these two characters off the boat?
- Dunkirk Character Stories
- Cómo expresar en español “comfort food”?
- How did so many Australian politicans win Parliamentary seats while unaware of or denying their second nationality?
- What does “Thou shalt not suffer a witch to live” mean?
- Found a handwritten note and I don't know the language
- Managing the bitcoin address gap
Deep Learning: What are the differences between DeepMind's Learning to Learn method and a grid search of a network's hyperparameters?
If we have a meta-learner that trains an optimizer (who contains certain hyperparameters) and the optimizer is fine-tuned by the meta-learner depending on how it performs, how is it different from a usual grid search of the best hyperparameters? One way I think it's different is that the meta-learner can supposedly 'intelligently' find the best optimizer but a gridsearch is rather a brute-force method. But a grid search would likely include human knowledge on the range of hyperparameters where the model is likely to perform well.
My current impression is that the meta-learner simply tweaks the hyperparameters of the optimizer (which are usually fixed in many other cases) after certain number of epochs, evaluates the performance, then dynamically changes how the hyperparameters should be tweaked. Is this what the authors of the paper did?