Making Machine Learning Usable

Framework for applying ML explanations and augmentations for real-world decision making

Github Join Slack community

Sibyl banner

Usable Machine Learning

Explain, Understand, Act

With the Sibyl libraries, you can augment machine learning applications for smoother human-ML interactions.

Sibyl Use Cases

Sibyl has been demonstrated in the following application domains.

Child welfare

Child welfare

Explaining the predictions of child welfare predictive risk models, to aid the referral screening process

Read the paper

Medicine

Medicine

Explaining surgery-outcome predictions on electronic health records

Read the paper

Software

Check back here soon as our libraries go public!

Get Involved

Sibyl Research

The Need for Interpretable Features: Motivation and Taxonomy

The Need for Interpretable Features: Motivation and Taxonomy

Alexandra Zytek, Ignacio Arnaldo, Dongyu Liu, Laure Berti-Equille, Kalyan Veeramachaneni

SIGKDD ACM SIGKDD Explorations Newsletter, 2022. (to appear)

Download: [doi] [pdf]

Sibyl: Understanding and Addressing the Usability Challenges of Machine Learning In High-Stakes Decision Making

Sibyl: Understanding and Addressing the Usability Challenges of Machine Learning In High-Stakes Decision Making

Alexandra Zytek, Dongyu Liu, Rhema Vaithianathan, Kalyan Veeramachaneni

TVCG IEEE Transactions on Visualization and Computer Graphics (VIS'21), 2022.

Download: [doi] [pdf] [bib] [video] [talk] [media]

VBridge: Connecting the Dots Between Features and Data to Explain Healthcare Models

VBridge: Connecting the Dots Between Features and Data to Explain Healthcare Models

Furui Cheng, Dongyu Liu, Fan Du, Yanna Lin, Alexandra Zytek, Haomin Li, Huamin Qu, Kalyan Veeramachaneni

TVCG IEEE Transactions on Visualization and Computer Graphics (VIS'21), 2022.

Download: [doi] [pdf] [bib] [video] [talk] | 🏆 Best Paper Honorable Mention (top 5% out of 440 submissions)

Sibyl: Explaining Machine Learning Models for Better Child Welfare Decision Making

Sibyl: Explaining Machine Learning Models for Better Child Welfare Decision Making

Alexandra Zytek, Dongyu Liu, Rhema Vaithianathan, and Kalyan Veeramachaneni

CHI EA ACM Conference on Human Factors in Computing Systems - Extended Abstracts, 2021.

Download: [doi] [pdf] [bib] [talk]

Team

ALEXANDRA ZYTEK

ALEXANDRA ZYTEK

PhD (Project Lead)

DAI Lab, MIT

DONGYU LIU

DONGYU LIU

Postdoc (Project Co-Lead)

DAI Lab, MIT

WARREN WANG

WARREN WANG

Undergraduate Researcher

DAI Lab, MIT

MICHAELA HENRY

MICHAELA HENRY

Program Coordinator

DAI Lab, MIT

Kalyan Veeramachaneni

Kalyan Veeramachaneni

Principal Investigator

DAI Lab, MIT

Design Team

SERGIU OJOC

SERGIU OJOC

Frontend Developer

DAI Lab, MIT

IULIA IONESCU

IULIA IONESCU

UI/UX Designer

DAI Lab, MIT

ARASH AKHGARI

ARASH AKHGARI

Animator & Graphic Designer

DAI Lab, MIT

Contact

Funding

This work is supported in part by NSF Award 1761812.

NSF logo