ÖAW AI Winter School 2023

23 Jan 2023, 08:30 → 27 Jan 2023, 17:00 Europe/Zurich

Seminar room 1 + 2 (Campus Akademie, Doktor-Ignaz-Seipel-Platz 2, 1010 Wien)

Wolfgang Waltenberger (Austrian Academy of Sciences (AT))

Welcome to the 3rd AI winter school of the Austrian Academy of Sciences (ÖAW), which is held on January 23 – 27, 2023 in Vienna. Here you can find some general information about the event, as well as the recordings and materials for the courses given at the winter school.

 

The topics of this installment are:

• introduction to natural language processing,
• sentiment analysis, toxicity detection,
• differentiable programming,
• simulation based inference,
• generative models and unsupervised learning.

 

The school is open to people affiliated with or working closely with institutes of the Austrian Academy of Sciences. All levels of seniority are welcome: students (MSc, PhD), PostDocs, faculty and staff.

 

A dedicated discord server may be used for instant communication.

 

Update: jan 30th: Thanks everybody for attending the school, here 

is a link to our feedback web form!

 

 

Local organizers:

• Thomas Astner (IQOQI)
• Peter Balazs (ARI)
• Brigitte DeMonte (HEPHY, secretary)
• Matej Durco (ACDH-CH)
• Nicki Holighaus (ARI)
• Alexandra Lenz (ACDH-CH)
• Daniel Schopper (ACDH-CH)
• Felix Wagner (HEPHY)
• Wolfgang Waltenberger (HEPHY, chair)
• Micheline Welte (ÖAI)

photos.zip

Monday, 23 January

08:30 → 09:00 Registration

Speaker: Brigitte De Monte

09:00 → 09:10 Welcome

Speaker: Wolfgang Waltenberger (Austrian Academy of Sciences (AT))

oeawai_intro.pdf

09:10 → 10:30 Introduction, NLP and deep learning

Speaker: Tomasz Walkowiak

Introduction, NLP and deep learning.pdf

10:30 → 11:00 Coffee Break

11:00 → 12:30 Introduction, Deep Learning and Natural Sciences

Speaker: Gregor Kasieczka (Hamburg University (DE))

DeepLearning_Vienna_OEAW_2023_Part_1.pdf

12:30 → 14:00 Lunch Break

14:00 → 15:30 Natural Language Processing

Speaker: Tomasz Walkowiak

Natural Language Processing.pptx.pdf

15:30 → 16:00 Coffee Break

16:00 → 18:00 Exercises

Speaker: Tomasz Walkowiak

[1] Simple classification

[2] BERT classification

18:00 → 20:00 Social Events: Welcome Reception

Tuesday, 24 January

09:00 → 10:30 Unsupervised methods

Speaker: Gregor Kasieczka (Hamburg University (DE))

DeepLearning_Vienna_OEAW_2023_Part_2.pdf

10:30 → 11:00 Coffee Break

11:00 → 12:30 Multidimensional Scaling, Data Visualization, and Explainable AI

Speakers: Mateusz Gniewkowski, Tomasz Walkowiak

Visualisation.pptx.pdf

12:30 → 14:00 Lunch Break

14:00 → 16:00 Exercises - Normalizing Flows

Speaker: Manuel Sommerhalder (Hamburg University (DE))

Flow Exercise

Flow Exercise Solution

16:00 → 16:30 Coffee Break

16:30 → 18:00 Exercises - Anomaly Detection

Speaker: Manuel Sommerhalder (Hamburg University (DE))

Anomaly Detection Exercise

Anomaly Detection Exercise Solution

Wednesday, 25 January

09:00 → 10:30 Sentiment Analysis

25 min Overview of Application of NLP in Recommender Systems and other areas + short introduction to Digital Humanism
45 min Sentiment Analysis
20 min Emotion Detection

Speakers: Julia Neidhardt, Mete Sertkan, Thomas Kolb

nlp_and_social_inpact_öaw-workshop-jneidhardt.pdf

sentiment_analysis_öaw-workshop-tkolb.pdf

tracing_the_evolution_of_emotion_studies_öaw-workshop-inalis-neuner.pdf

10:30 → 11:00 Coffee Break

11:00 → 12:30 Toxicity, and Populism

25 min Emotion Detection
25 min Psychological perspective
15 min Toxicity Detection
25 min Populism Detection

Speakers: Ahmadou Wagne, Irina Nalis, Julia Neidhardt

populism_detection_öaw-workshop-awagne.pdf

12:30 → 14:00 Lunch Break

14:00 → 16:00 Exercises

https://github.com/CDL-RecSys/oeaw-ai-winter-school-2023

Speakers: Julia Neidhardt, Meta Sertkan, Thomas Kolb

16:00 → 16:30 Coffee Break

16:30 → 18:00 Exercises

Speakers: Adhmadou Wagne, Irina Nalis, Julia Neidhardt

18:00 → 20:30 Social Events: Hackathon

Thursday, 26 January

09:00 → 10:30 Differentiable Programming

Speaker: Lukas Alexander Heinrich (Technische Universitat Munchen (DE))

OAW_WinterSchool_2023_AD.pdf

10:30 → 11:00 Coffee Break

11:00 → 12:30 Simulation Based Inference

Speaker: Lukas Alexander Heinrich (Max Planck Society (DE))

OAW_WinterSchool_2023_SBI.pdf

12:30 → 14:00 Lunch Break

14:00 → 16:00 Exercises

Speaker: Lukas Alexander Heinrich (Max Planck Society (DE))

AttentionNotebook

Autodiff Tutorial

16:00 → 16:30 Coffee Break

16:30 → 18:00 Exercises

Speaker: Lukas Alexander Heinrich (Max Planck Society (DE))

18:30 → 21:30 Social Events: Social Dinner

Friday, 27 January

09:30 → 11:00 Exercises

Speaker: Tomasz Walkowiak

Visualisation

11:00 → 11:30 Coffee Break

11:30 → 12:50 Exercises - Model-Agnostic Explanation Techniques

Speaker: Mateusz Gniewkowski

XAI

12:50 → 13:00 Closing

Speaker: Wolfgang Waltenberger (Austrian Academy of Sciences (AT))

oeawai_outro.pdf