Daniela KeusenJan 21, 2024

#KünstlicheIntelligenz und #MaschinellesLernen in der Praxis
#MOOC im Rahmen der #KIwinterschool auf #openHPI @Hasso_Plattner_Institute

#KI-/ #ML-Projekte anhand #CRISP-#DM Prozess
#MachineLearning

#Python #Programmierbibliotheken #Datenvisualisierung 
#RecommenderSysteme #Datenvorverarbeitung
#Datenformate #Datenrepräsentationen #Hyperparameter #SentimentAnalyse #Datenlabeling #LargeScaleLanguageModels
#ComputerVision #Datenbeschaffung #Explainability #Bias

https://open.hpi.de/courses/kipraxis2021

Science SlamSep 21, 2023

🧐 Künstlich Intelligent lernen 💻 – Theresa Eimer beschäftigt sich als Informatikerin damit, wie man künstlich intelligent lernen kann. Besonders wichtig dabei: Hyperparameter. Doch wie funktioniert maschinelles Lernen überhaupt?🎓
Hier gehts zum Vortrag: https://youtu.be/UQ4y5ovvp2Y

#forschung #wissenschaft #scienceslam #science #lernen #study #AI #informatik #maschinelleslernen #Hyperparameter #künstlicheintelligenz #daten #datenkrake #blackbox

Wie du künstlich intelligentes Lernen richtig umsetzt (Theresa Eimer - Science Slam)

YouTube
New Submissions to TMLRJun 8, 2023

Sample Average Approximation for Black-Box Variational Inference

https://openreview.net/forum?id=Lvg10LZ5nL

#variational #optimization #hyperparameter

Sample Average Approximation for Black-Box Variational Inference

We present a novel approach for black-box VI that bypasses the difficulties of stochastic gradient ascent, including the task of selecting step-sizes. Our approach involves using a sequence of...

OpenReview
Published papers at TMLRFeb 2, 2023

No More Pesky Hyperparameters: Offline Hyperparameter Tuning for RL

Han Wang, Archit Sakhadeo, Adam M White et al.

https://openreview.net/forum?id=AiOUi3440V

#hyperparameters #hyperparameter #learns

No More Pesky Hyperparameters: Offline Hyperparameter Tuning for RL

The performance of reinforcement learning (RL) agents is sensitive to the choice of hyperparameters. In real-world settings like robotics or industrial control systems, however, testing different...

OpenReview
Show threadBlake RichardsNov 29, 2022

2/10) #SSL models show great promise and can learn #representations from large-scale unlabelled data. But, identifying the best model across diff #hyperparameter configs requires measuring downstream task performance, which requires #labels and adds to the #compute time+resources. 😕

#AI #ML #deeplearning

maxyNov 9, 2022

I wonder if anyone has written about #ASHA #hyperparameter optimization vs restart strategies.

I've been playing around with ASHA in #ray on my CMA-ES optimization task. Turns out my hyperparameters (population_size, sigma_0) are almost uncorrelated to performance, but the final performance varies a lot per run.

So what ASHA actually does for me is just picking the lucky runs. Which helped a lot. Maybe I should try this BIPOP restarting strategy of CMA-ES instead.