[HN Gopher] AutoML-Zero: Evolving Machine Learning Algorithms fr... ___________________________________________________________________ AutoML-Zero: Evolving Machine Learning Algorithms from Scratch Author : lainon Score : 69 points Date : 2020-03-10 20:08 UTC (2 hours ago) (HTM) web link (github.com) (TXT) w3m dump (github.com) | joe_the_user wrote: | _AutoML-Zero aims to automatically discover computer programs | that can solve machine learning tasks, starting from empty or | random programs and using only basic math operations._ | | If this system is not using human bias, who is it choosing what | good program is? Surely, human labeling data involves humans | adding their bias to the data? | | It seems like AlphaGoZero was able to do just end-to-end ML | because it was able to use a very clear and "objective" standard, | whether a program wins or loses at the game of Go. | | Would this approach only deal with similarly unambiguous | problems? | | Edit: also, AlphaGoZero was one of the most ML ever created (at | least at the time of its creation). How much computing resources | would this require for more fully general learning? Will there be | a limit to such an approach? | darawk wrote: | > It seems like AlphaGoZero was able to do just end-to-end ML | because it was able to use a very clear and "objective" | standard, whether a program wins or loses at the game of Go. | | Just a fun note: winning or losing at the game of Go is | actually surprisingly subjective: | | https://en.wikipedia.org/wiki/Go_(game)#Scoring_rules | jxcole wrote: | Interesting, but how does it perform on standard benchmarks like | image net and MNIST? | p1esk wrote: | They have some cifar10 results in the paper, but only very | small networks. | lokimedes wrote: | Reminds me of https://www.nutonian.com/products/eureqa/ which I | used quite productively to model multivariate distributions from | data back in the 2000's. Funny how everything stays the same, but | with a new set of players on the bandwagon. | TaylorAlexander wrote: | Shouldn't this link directly to the Readme? | | https://github.com/google-research/google-research/blob/mast... | manually wrote: | Next: | | - Autosuggest database tables to use | | - Automatically reserve parallel computing resources | | - Autodetect data health issues and auto fix them | | - Autodetect concept drift and auto fix it | | - Auto engineer features and interactions | | - Autodetect leakage and fix it | | - Autodetect unfairness and auto fix it | | - Autocreate more weakly-labelled training data | | - Autocreate descriptive statistics and model eval stats | | - Autocreate monitoring | | - Autocreate regulations reports | | - Autocreate a data infra pipeline | | - Autocreate a prediction serving endpoint | | - Auto setup a meeting with relevant stakeholders on Google | Calendar | | - Auto deploy on Google Cloud | | - Automatically buy carbon offset | | - Auto fire your in-house data scientists | neximo64 wrote: | Would be funny but most of those things are already on AutoML | Tables, including the carbon offset | | https://cloud.google.com/automl-tables | otagekki wrote: | Poor data scientists, now whose heads get cut when things go | wrong and companies lose billions? | manually wrote: | In the days when Sussman was a novice, Minsky once came to | him as he sat hacking at the PDP-6. | | "What are you doing?", asked Minsky. | | "I am training a randomly wired neural net to play Tic-Tac- | Toe" Sussman replied. | | "Why is the net wired randomly?", asked Minsky. | | "I do not want it to have any preconceptions of how to play", | Sussman said. | | Minsky then shut his eyes. | | "Why do you close your eyes?", Sussman asked his teacher. | | "So that the room will be empty." | | At that moment, Sussman was enlightened. | xiaodai wrote: | Autodetect data health issues and auto fix them | | Funy you say that cos my company is actually developing | something along those lines ___________________________________________________________________ (page generated 2020-03-10 23:00 UTC)