Data Analytics Perspectives: Machine Learning

Neil D. Lawrence

Data Analytics Perspectives: Machine Learning

CSaP Annual Conference, Royal Society

29th June 2017

Neil Lawrence

Amazon Research Cambridge and University of Sheffield

@lawrennd inverseprobability.com

What is Machine Learning?

data + model + compute -> prediction

Royal Society Report

Machine Learning: Power and Promise of Computers that Learn by Example

Two Phenomena underpinned by ML

  • Data Science

  • Artificial Intelligence

Operating at Different Time Scales

data science  artificial intelligence

Data Science

Data Science

  • New technologies historically led to new professions:

    • Brunel (born 1806): Civil, mechanical, naval

    • Tesla (born 1856): Electrical and power

    • William Shockley (born 1910): Electronic

    • Watts S. Humphrey (born 1927): Software

The Software Crisis

The major cause of the software crisis is that the machines have become several orders of magnitude more powerful! To put it quite bluntly: as long as there were no machines, programming was no problem at all; when we had a few weak computers, programming became a mild problem, and now we have gigantic computers, programming has become an equally gigantic problem.

Edsger Dijkstra, The Humble Programmer

The Data Crisis

The major cause of the data crisis is that machines have become more interconnected than ever before. Data access is therefore cheap, but data quality is often poor or personally sensitive. We need cheap high quality data and systems which protect individual privacy.

Me (born 1972)

Artificial Intelligence

AI Bubble?

AI Bubble?

AI Bubble?

Artificial Intelligence

  • Challenge of empathy.

“AlphaGo will replace accountants next”

BEIS Discussion Under Chatham House Rule

Artificial Intelligence

  • Challenge of empathy.

“It doesn’t even replace a human Go player!”

Thinks me

Deploying ML in Real World: Machine Learning Systems Design

  • Internet of Things

  • Major new challenge for systems designers.

  • AI systems are currently fragile

  • Example: Stuxnet

Deploying ML in Real World: Machine Learning Systems Design

  • Internet of Things

  • Major new challenge for systems designers.

  • AI systems are currently fragile

Deploying ML in Real World: Machine Learning Systems Design

  • Internet of People

  • Major new challenge for systems designers.

  • AI systems are currently fragile

Machine Learning Systems Design

Peppercorns

  • A new name for system failures which aren’t bugs.

  • Difference between finding a fly in your soup vs a peppercorn in your soup.

Thanks!