Human-Centered Machine Learning
Saarland University Advanced Course, WS 2018
Tentative Schedule
Tentative schedule for lectures, exams and assignment deadlines (all lectures held in room 005, building E 1.5 in SB and room 113, building G26 in KL). All information is subject to change.
- 16/10 (Lec 1): Introduction
- 18/10 (Lec 2): Notions and Measures for Discrimination
- Lecture Notes: Lec 2
- Suggested Reading: Big Data's Disparate Impact. Barocas, S., & Selbst, A. D. California Law Review 2016
- Suggested Reading: Fair prediction with disparate impact: A study of bias in recidivism prediction instruments. Chouldechova, A. Big Data 2017
- 23/10 (Lec 3): Detecting Discrimination
- Lecture Notes: Lec 3
- Reading Assignment (Submission Deadline: 23/10, 11:59 AM): Discrimination-aware Data Mining. Pedreshi, D., Ruggieri, S., & Turini, F. SIGKDD 2008
- Suggested Reading: From Parity to Preference-based Notions of Fairness in Classification. Zafar, M. B., Valera, I., Rodriguez, M., Gummadi, K., & Weller, A. NIPS 2017
- 25/10 (Lec 4): Detecting Discrimination
- Lecture Notes: Lec 4
- Reading Assignment (Submission Deadline: 25/10, 11:59 AM): k-NN as an implementation of situation testing for discrimination discovery and prevention. Luong, B. T., Ruggieri, S., & Turini, F. SIGKDD 2011
- 30/10 (Lec 5): Mitigating Discrimination: In-processing Methods
- Lecture Notes: Lec 5
- Reading Assignment (Submission Deadline: 30/10, 11:59 AM): Fairness Beyond Disparate Treatment & Disparate Impact: Learning Classification without Disparate Mistreatment. Zafar, M. B., Valera, I., Gomez Rodriguez, M., & Gummadi, K. P. WWW 2017
- 01/11 (Bank holiday): -
- 06/11 (Lec 6): Mitigating Discrimination: Post-processing Methods
- Reading Assignment (Submission Deadline: 06/11, 11:59 AM): Equality of Opportunity in Supervised Learning. Hardt, M., Price, E., & Srebro, N. NIPS 2016
- 08/11 (Lec 7): Mitigating Discrimination: Pre-processing Methods
- Reading Assignment (Submission Deadline: 08/11, 11:59 AM): Learning Fair Representations. Zemel, R., Wu, Y., Swersky, K., Pitassi, T., & Dwork, C. ICML 2013
- 13/11 (Lec 8): Individual Unfairness: Social Welfare and Inequality Measurement
- Reading Assignment (Submission Deadline: 13/11, 11:59 AM): On the Measurement of Inequality. Atkinson, A. B. Journal of economic theory 1970
- 15/11 (Lec 9): Individual vs. Group Unfairness
- Lecture Notes: Lec 9
- Reading Assignment (Submission Deadline: 15/11, 11:59 AM): A Unified Approach to Quantifying Algorithmic Unfairness: Measuring Individual & Group Unfairness via Inequality Indices. Speicher, T., Heidari, H., Grgic-Hlaca, N., Gummadi, K. P., Singla, A., Weller, A., & Zafar, M. B. KDD 2018
- 20/11 (Lec 10): Bias in Search and Recommender Systems
- Reading Assignment (Submission Deadline: 20/11, 11:59 AM): Man is to computer programmer as woman is to homemaker? debiasing word embeddings. Bolukbasi, T., Chang, K. W., Zou, J. Y., Saligrama, V., & Kalai, A. T. NIPS 2016
- 22/11 (Lec 11): Bias in Search and Recommender Systems
- Lecture Notes: Lec 11
- Reading Assignment (Submission Deadline: 22/11, 11:59 AM): Automated experiments on ad privacy settings. Datta, A., Tschantz, M. C., & Datta, A. PETS 2015
- 27/11 (Lec 12): Procedural Fairness
- Lecture Notes: Lec 12
- Reading Assignment (Submission Deadline: 27/11, 11:59 AM): Human Perceptions of Fairness in Algorithmic Decision Making: A Case Study of Criminal Risk Prediction. Grgic-Hlaca, N., Redmiles, E. M., Gummadi, K. P., & Weller, A. WWW 2018
- Suggested Reading: Beyond Distributive Fairness in Algorithmic Decision Making: Feature Selection for Procedurally Fair Learning. Grgic-Hlaca, N., Zafar, M. B., Gummadi, K. P., & Weller, A. AAAI 2018
- 29/11 (Lec 13): Informational Fairness
- Reading Assignment (Submission Deadline: 29/11, 11:59 AM): Counterfactual explanations without opening the black box: Automated decisions and the GDPR. Wachter, S., Mittelstadt, B., & Russell, C. 2017
- 04/12 (Review):
- 06/12 (Midterm):
- 11/12 (Lec 14): Introduction to temporal point processes (I)
- 13/12 (Lec 15): Introduction to temporal point processes (II)
- Suggested Reading: On Lewis' Simulation Method for Point Processes. Ogata, Y. IEEE Transactions on Information theory, vol 27 (1), 1981
- 18/12 (Lec 16): Advanced concepts in temporal point processes
- Lecture notes: pdf, pptx
- Notes on Temporal Point Processes
- Suggested Reading: The Neural Hawkes Process: A Neurally Self-Modulating Multivariate Point Process. Mei, H., & Eisner, J. NIPS 2017
- 20/12 (Lec 17): First Coding Assignment
- Lecture notes: pdf, pptx
- Source code
- Lecture recording (700Mb): Video
- 08/01 (Lec 18): Information Propagation
- Lecture notes: pdf, pptx
- Reading Assignment (Submission Deadline: 08/01, 11:59 AM): COEVOLVE: A Joint Point Process Model for Information Diffusion and Network Evolution. Farajtabar, M., Wang, Y., Gomez-Rodriguez, M., Li, S., Zha, H., & Song, L. Journal of Machine Learning Research 18(41):1−49, 2017.
- 10/01 (Lec 19): Opinion Dynamics
- Lecture notes: pdf, pptx
- Reading Assignment (Submission Deadline: 10/01, 11:59 AM): Learning and Forecasting Opinion Dynamics in Social Networks. De, A., Valera, I., Ganguly, N., Bhattacharya, S., & Gomez-Rodriguez, M. NIPS 2016
- 15/01 (Lec 20): Information Reliability
- Lecture notes: pdf, pptx
- Reading Assignment (Submission Deadline: 15/01, 11:59 AM): Distilling Information Reliability and Source Trustworthiness from Digital Traces. Tabibian, B., Valera, I., Farajtabar, M., Song, L., Schölkopf, B., & Gomez-Rodriguez, M. WWW 2017
- 17/01 (Lec 21): Optimal Control of TPPs
- 22/01 (Lec 22): Viral Marketing
- Lecture notes: pdf, pptx (optional)
- Reading Assignment (Submission Deadline: 22/01, 11:59 AM): RedQueen: An Online Algorithm for Smart Broadcasting in Social Networks. A. Zarezade, U. Upadhyay, H. Rabiee & Gomez-Rodriguez, M. WSDM 2016
- 24/01 (Lec 23): Second Coding Assignment
- Lecture notes: pdf, pptx
- Source code
- Lecture recording (500Mb): Video
- 29/01 (Lec 24): Reinforcement Learning of TPPs
- Lecture notes: pdf, pptx
- Reading Assignment (Submission Deadline: 29/01, 11:59 AM): Deep Reinforcement Learning of Marked Temporal Point Processes. U. Upadhyay, A. De & M. Gomez-Rodriguez NeurIPS 2018
- 31/01 (Lec 25): Enhancing Human Learning
- 05/02 (Review):
- 07/02 (Final):
