Re:Inverted Transformers Are Effective for Time Series Forec...
by Nevermark on Wednesday October 11, 2023 @06:27PM
(#37851141)
Deep learning models are not better than vast problem areas with analytical design algorithms. Deep learning's succession of many triumphs is across areas where analytical design has proven difficult.
There are many optimal, or known near optimal, direct design algorithms for systems that are well characterized. Not only do you get a solution with guaranteed qualities, but any analysis, including verification of reliability and accuracy, and valid operating regimes, of such solutions go way beyond what deep learning models provide.
Second, just assuming that a new algorithm is better than a previous algorithm, without testing, is completely irrational. An anathema to the entire purpose and benefits of science, math, and rational research in general.
If you are going to propose switching algorithms for a long known class of problems, you need to measure the new algorithm against the previous state of the art.
Otherwise people will drive into unnecessary deadends, deploy pointlessly bad designs, forget important knowledge that still has value, and worst of all, lose out on what the older algorithms can suggest for improving newer algorithms, with no excuse but carelessness.
There are many optimal, or known near optimal, direct design algorithms for systems that are well characterized. Not only do you get a solution with guaranteed qualities, but any analysis, including verification of reliability and accuracy, and valid operating regimes, of such solutions go way beyond what deep learning models provide.
Second, just assuming that a new algorithm is better than a previous algorithm, without testing, is completely irrational. An anathema to the entire purpose and benefits of science, math, and rational research in general.
If you are going to propose switching algorithms for a long known class of problems, you need to measure the new algorithm against the previous state of the art.
Otherwise people will drive into unnecessary deadends, deploy pointlessly bad designs, forget important knowledge that still has value, and worst of all, lose out on what the older algorithms can suggest for improving newer algorithms, with no excuse but carelessness.