Post-Digital Transformation: Intellectual Debt
Cambridge Senior Management Programme, Judge Business School, University of Cambridge
Introduction
The Gartner Hype Cycle
Cycle for ML Terms
Google Trends
What is Machine Learning?
What is Machine Learning?
\[ \text{data} + \text{model} \stackrel{\text{compute}}{\rightarrow} \text{prediction}\]
- data : observations, could be actively or passively acquired (meta-data).
- model : assumptions, based on previous experience (other data! transfer learning etc), or beliefs about the regularities of the universe. Inductive bias.
- prediction : an action to be taken or a categorization or a quality score.
- Royal Society Report: Machine Learning: Power and Promise of Computers that Learn by Example
What is Machine Learning?
\[\text{data} + \text{model} \stackrel{\text{compute}}{\rightarrow} \text{prediction}\]
- To combine data with a model need:
- a prediction function \(f(\cdot)\) includes our beliefs about the regularities of the universe
- an objective function \(E(\cdot)\) defines the cost of misprediction.
Artificial Intelligence and Data Science
- AI aims to equip computers with human capabilities
- Image understanding
- Computer vision
- Speech recognition
- Natural language understanding
- Machine translation
Supervised Learning for AI
- Dominant approach today:
- Generate large labelled data set from humans.
- Use supervised learning to emulate that data.
- E.g. ImageNet Russakovsky et al. (2015)
- Significant advances due to deep learning
- E.g. Alexa, Amazon Go
Data Science
- Arises from happenstance data.
- Differs from statistics in that the question comes after data collection.
Exercise: Score Yourself (in chat)
- I am a data science:
- follower (no visibility/influence)
- some visibilty/influence
- visibility and some influence
- leader (lead on data and AI developments)
Intellectual Debt
Information and Embodiment
Embodiment Factors
| bits/min | billions | 2,000 |
|
billion calculations/s |
~100 | a billion |
| embodiment | 20 minutes | 5 billion years |
Evolved Relationship with Information
New Flow of Information
Evolved Relationship
Evolved Relationship
There are three types of lies: lies, damned lies and statistics
??
There are three types of lies: lies, damned lies and statistics
Benjamin Disraeli
There are three types of lies: lies, damned lies and statistics
Benjamin Disraeli 1804-1881
There are three types of lies: lies, damned lies and ‘big data’
Neil Lawrence 1972-?
Mathematical Statistics
‘Mathematical Data Science’
Heider and Simmel (1944)
The Big Data Paradox
- We collect more data, but we understand less.
Wood or Tree
Big Model Paradox
- Add complexity to the model to make it realistic.
- Move model “beyond human intuition”
- But model still falls well short of mark in terms of representing reality
Complexity in Action
Data Selective Attention Bias
BMI Steps Data
BMI Steps Data Analysis
A Hypothesis as a Liability
“ ‘When someone seeks,’ said Siddhartha, ‘then it easily happens that his eyes see only the thing that he seeks, and he is able to find nothing, to take in nothing. […] Seeking means: having a goal. But finding means: being free, being open, having no goal.’ ”
Hermann Hesse
The Scientific Process
Number Theatre
Data Theatre
The Art of Statistics
Increasing Need for Human Judgment
The domain of human judgment is increasing.
How these firms use knowledge. How do they generate ideas?
Case Study: 10:30
15 Minute Break!
Case Study: Face Masks
DELVE Overview
DELVE will contribute data driven analysis to complement the evidence base informing the UK’s strategic response, by:
- Analysing national and international data to determine the effect of different measures and strategies on a range of public health, social and economic outcomes
- Using emerging sources of data as new evidence from the unfolding pandemic comes to light
- Ensuring that the work of this group is coordinated with others and communicated as necessary both nationally and internationally
Delve Timeline
- First contact 3rd April
- First meeting 7th April
- First working group 16th April
Data at the Heart
- Use data to answer policy questions.
- Make international comparisons for input.
- Challenges: around getting data.
Breakout
- As pre-read material, you were provided with a report summary and links to the full report and responses collated by the Science Media Centre.
- In your groups, discuss the summary and these responses.
- For both the comments and the responses put yourself in the position of a government minister.
- What would you do and why?
- Which comments and recommendations were helpful and which weren’t?}
Links
Bringing it Back
Dealing with Intellectual Debt
What we Did at Amazon
Are Right a Lot
Leaders are right a lot. They have strong judgment and good instincts. They seek diverse perspectives and work to disconfirm their beliefs.
Dive Deep
Leaders operate at all levels, stay connected to the details, audit frequently, and are skeptical when metrics and anecdote differ. No task is beneath them.
Thoughtsday
- Dealing with highly operational decision making.
What we did in DELVE
Data as a Convener
- Data allows externalisation of cognition.
- Even when not existing, can ask: What data would we want?
Breakout: What would you do in your company?
- Finish 12:30