ethical machines: data mining and fairness – the optimistic view
TRANSCRIPT
![Page 1: Ethical machines: data mining and fairness – the optimistic view](https://reader031.vdocument.in/reader031/viewer/2022030308/58f0ab791a28abeb188b45f9/html5/thumbnails/1.jpg)
Ethical machines: data mining and fairness – the optimistic view Anna Ronkainen chief scientist, TrademarkNow it’s complicated, UU of Helsinki & Turku @ ronkaine 2016-05-02
![Page 2: Ethical machines: data mining and fairness – the optimistic view](https://reader031.vdocument.in/reader031/viewer/2022030308/58f0ab791a28abeb188b45f9/html5/thumbnails/2.jpg)
My three points 1. people aren’t exactly perfect, either, and
sometimes algorithms can be an improvement
2. different types of algorithms needed for arriving at decisions and validating/disproving them
3. data protection law about automated decision-making needs to be taken seriously
![Page 3: Ethical machines: data mining and fairness – the optimistic view](https://reader031.vdocument.in/reader031/viewer/2022030308/58f0ab791a28abeb188b45f9/html5/thumbnails/3.jpg)
Heuristics or biases?
(Dhami 2003)
![Page 4: Ethical machines: data mining and fairness – the optimistic view](https://reader031.vdocument.in/reader031/viewer/2022030308/58f0ab791a28abeb188b45f9/html5/thumbnails/4.jpg)
Sometimes people fail in unexpected ways...
(Danziger et al (2011):Extraneous Factors in Judicial Decisions)
![Page 5: Ethical machines: data mining and fairness – the optimistic view](https://reader031.vdocument.in/reader031/viewer/2022030308/58f0ab791a28abeb188b45f9/html5/thumbnails/5.jpg)
Systems 1 and 2 in legal reasoning: interaction System 1: making the decision System 2: validation and justification
(Ronkainen2011)
![Page 6: Ethical machines: data mining and fairness – the optimistic view](https://reader031.vdocument.in/reader031/viewer/2022030308/58f0ab791a28abeb188b45f9/html5/thumbnails/6.jpg)
Implications for algorithms (hypothesis) - System-1-like processes cannot be captured
reliably with GOFAI -> machine learning and other statistical approaches needed
- the System 2 part (finding supporting arguments and validating/falsifying the decision candidate) can (and should) be implemented with rule-based GOFAI for accountability, maintainability etc etc etc
![Page 7: Ethical machines: data mining and fairness – the optimistic view](https://reader031.vdocument.in/reader031/viewer/2022030308/58f0ab791a28abeb188b45f9/html5/thumbnails/7.jpg)
Taking data protection seriously?
(2016 EU General Data Protection Regulation)
![Page 8: Ethical machines: data mining and fairness – the optimistic view](https://reader031.vdocument.in/reader031/viewer/2022030308/58f0ab791a28abeb188b45f9/html5/thumbnails/8.jpg)
Seriously-seriously?
(1995 EU Data Protection Directive 95/46/EC)
![Page 9: Ethical machines: data mining and fairness – the optimistic view](https://reader031.vdocument.in/reader031/viewer/2022030308/58f0ab791a28abeb188b45f9/html5/thumbnails/9.jpg)
My three points 1. people aren’t exactly perfect, either, and
sometimes algorithms can be an improvement
2. different types of algorithms needed for arriving at decisions and validating/disproving them
3. data protection law about automated decision-making needs to be taken seriously
![Page 10: Ethical machines: data mining and fairness – the optimistic view](https://reader031.vdocument.in/reader031/viewer/2022030308/58f0ab791a28abeb188b45f9/html5/thumbnails/10.jpg)
Thank you!