TopConf Linz, 02/02/2016
0 likes792 views
This document discusses the relationship between probability, machine learning, and user experience in consumer IoT devices. It notes that individual apps are not enough and that intelligent devices need to have contextual awareness and conversational abilities based on things like schedule, location, and preferences. However, acquiring and applying knowledge through machine learning comes with challenges around explaining probabilistic concepts and predictions to users, and ensuring that conversational systems do not feel repetitive, disruptive, or intrusive. Quantitative thinking will become more important as life becomes more dependent on probabilities modeled through data and algorithms.
![health
management
air conditioning
smart heating
communications
security
entertainment
lighting controlweather
monitoring
room occupancy
individual apps are NOT [ðə ˈfjuːʧə]](https://image.slidesharecdn.com/topconflinz020216-160205130731/85/TopConf-Linz-02-02-2016-6-320.jpg)
![30 40 50 60 70
average speed at this point [MPH]
time to target
[min]
10
20
30
40
50
we have a
sense for
simple
probabilities](https://image.slidesharecdn.com/topconflinz020216-160205130731/85/TopConf-Linz-02-02-2016-14-320.jpg)
TopConf Linz, 02/02/2016
- 1. the friction zone between probability, machine learning and user experience in the consumer IoT @BorisAdryan
- 4. 2016: a little reality check aka
- 5. modified, image from http://www.householdappliancesworld.com health management air conditioning smart heating communications security entertainment lighting controlweather monitoring room occupancy
- 6. health management air conditioning smart heating communications security entertainment lighting controlweather monitoring room occupancy individual apps are NOT [ðə ˈfjuːʧə]
- 7. source lost, seen on Twitter we want intelligent things that talk
- 8. David Rose: enchanted objects as UX paradigm modified, image from http://enchantedobjects.com
- 11. acquiring knowledge == learning machine learning creative thinking != decision making with statistics + algorithms ==
- 12. raw data information knowledge actionable insight action reaction barometric pressure, temperature, coordinates, schedule, … snow storm coming, airport hotel, need to travel flying and snow don’t go together rebook flight “context” structure rules “conversational” dynamic acquired
- 13. there’s no absolute truth out there data ✓ hard facts ✓ intuitive probability ✓ likelihood of some hypothesis being true given the data
- 14. 30 40 50 60 70 average speed at this point [MPH] time to target [min] 10 20 30 40 50 we have a sense for simple probabilities
- 15. who likes more detail? aka
- 16. p-value < 2.2e-16 confidence interval FDR posterior probability 95% 0.025 5% explain this to your neighbour Battery is goingto die today, p < 2.7x10-3
- 17. simple x -> y mapping x k h l q w g f -> y mapping your FitBit temperature your friend’s dog the car computationally, your life is incredibly messy
- 18. blog post at https://iot.ghost.io/is-it-all-machine-learning
- 19. data temperature wind speed wind direction precipitation air pressure airport code airline aircraft fully booked? avg delays cancellations serve booze? black box training flights cancelled in the past classifier ranked list of relevant features weight of features thresholds for features performance metric new data prediction
- 20. the hypothesis itself is a mathematical model explain this to your user
- 21. good decisions are based on experience machine learning is an iterative process training classifier performance assessment good enough? get on with life moredatafortraining data no yes
- 23. the issue with missing data given all features, we can discover the causality between them
- 24. self-learning systems will have to seek ‘missing’ data other than saying ‘urgent meeting’ in the calendar, how can the system know it’s really urgent? …preemptively
- 26. it’s none of your effing business
- 27. things getting more creepy… “Is there something you should tell me, Boris? I thought your wife was travelling…” …when they’re conversational
- 28. life is becoming increasingly dependent on probabilities and abstract quantities @BorisAdryan adding to our anxiety of uncertainty, the conversational IoT may potentially feel repetitive, disruptive and intrusive! quantitative and computational thinking is going to become an essential skill