Machine Learning Systems Design

[edit]

at Re-work Deep Learning Summit on Sep 20, 2019 [jupyter][reveal]
Neil D. Lawrence, University of Sheffield

Abstract

Machine learning solutions, in particular those based on deep learning methods, form an underpinning of the current revolution in “artificial intelligence” that has dominated popular press headlines and is having a significant influence on the wider tech agenda. In this talk I will give an overview of where we are now with machine learning solutions, and what challenges we face both in the near and far future. These include practical application of existing algorithms in the face of the need to explain decision-making, mechanisms for improving the quality and availability of data, dealing with large unstructured datasets.

$$\newcommand{\tk}[1]{} %\newcommand{\tk}[1]{\textbf{TK}: #1} \newcommand{\Amatrix}{\mathbf{A}} \newcommand{\KL}[2]{\text{KL}\left( #1\,\|\,#2 \right)} \newcommand{\Kaast}{\kernelMatrix_{\mathbf{ \ast}\mathbf{ \ast}}} \newcommand{\Kastu}{\kernelMatrix_{\mathbf{ \ast} \inducingVector}} \newcommand{\Kff}{\kernelMatrix_{\mappingFunctionVector \mappingFunctionVector}} \newcommand{\Kfu}{\kernelMatrix_{\mappingFunctionVector \inducingVector}} \newcommand{\Kuast}{\kernelMatrix_{\inducingVector \bf\ast}} \newcommand{\Kuf}{\kernelMatrix_{\inducingVector \mappingFunctionVector}} \newcommand{\Kuu}{\kernelMatrix_{\inducingVector \inducingVector}} \newcommand{\Kuui}{\Kuu^{-1}} \newcommand{\Qaast}{\mathbf{Q}_{\bf \ast \ast}} \newcommand{\Qastf}{\mathbf{Q}_{\ast \mappingFunction}} \newcommand{\Qfast}{\mathbf{Q}_{\mappingFunctionVector \bf \ast}} \newcommand{\Qff}{\mathbf{Q}_{\mappingFunctionVector \mappingFunctionVector}} \newcommand{\aMatrix}{\mathbf{A}} \newcommand{\aScalar}{a} \newcommand{\aVector}{\mathbf{a}} \newcommand{\acceleration}{a} \newcommand{\bMatrix}{\mathbf{B}} \newcommand{\bScalar}{b} \newcommand{\bVector}{\mathbf{b}} \newcommand{\basisFunc}{\phi} \newcommand{\basisFuncVector}{\boldsymbol{ \basisFunc}} \newcommand{\basisFunction}{\phi} \newcommand{\basisLocation}{\mu} \newcommand{\basisMatrix}{\boldsymbol{ \Phi}} \newcommand{\basisScalar}{\basisFunction} \newcommand{\basisVector}{\boldsymbol{ \basisFunction}} \newcommand{\activationFunction}{\phi} \newcommand{\activationMatrix}{\boldsymbol{ \Phi}} \newcommand{\activationScalar}{\basisFunction} \newcommand{\activationVector}{\boldsymbol{ \basisFunction}} \newcommand{\bigO}{\mathcal{O}} \newcommand{\binomProb}{\pi} \newcommand{\cMatrix}{\mathbf{C}} \newcommand{\cbasisMatrix}{\hat{\boldsymbol{ \Phi}}} \newcommand{\cdataMatrix}{\hat{\dataMatrix}} \newcommand{\cdataScalar}{\hat{\dataScalar}} \newcommand{\cdataVector}{\hat{\dataVector}} \newcommand{\centeredKernelMatrix}{\mathbf{ \MakeUppercase{\centeredKernelScalar}}} \newcommand{\centeredKernelScalar}{b} \newcommand{\centeredKernelVector}{\centeredKernelScalar} \newcommand{\centeringMatrix}{\mathbf{H}} \newcommand{\chiSquaredDist}[2]{\chi_{#1}^{2}\left(#2\right)} \newcommand{\chiSquaredSamp}[1]{\chi_{#1}^{2}} \newcommand{\conditionalCovariance}{\boldsymbol{ \Sigma}} \newcommand{\coregionalizationMatrix}{\mathbf{B}} \newcommand{\coregionalizationScalar}{b} \newcommand{\coregionalizationVector}{\mathbf{ \coregionalizationScalar}} \newcommand{\covDist}[2]{\text{cov}_{#2}\left(#1\right)} \newcommand{\covSamp}[1]{\text{cov}\left(#1\right)} \newcommand{\covarianceScalar}{c} \newcommand{\covarianceVector}{\mathbf{ \covarianceScalar}} \newcommand{\covarianceMatrix}{\mathbf{C}} \newcommand{\covarianceMatrixTwo}{\boldsymbol{ \Sigma}} \newcommand{\croupierScalar}{s} \newcommand{\croupierVector}{\mathbf{ \croupierScalar}} \newcommand{\croupierMatrix}{\mathbf{ \MakeUppercase{\croupierScalar}}} \newcommand{\dataDim}{p} \newcommand{\dataIndex}{i} \newcommand{\dataIndexTwo}{j} \newcommand{\dataMatrix}{\mathbf{Y}} \newcommand{\dataScalar}{y} \newcommand{\dataSet}{\mathcal{D}} \newcommand{\dataStd}{\sigma} \newcommand{\dataVector}{\mathbf{ \dataScalar}} \newcommand{\decayRate}{d} \newcommand{\degreeMatrix}{\mathbf{ \MakeUppercase{\degreeScalar}}} \newcommand{\degreeScalar}{d} \newcommand{\degreeVector}{\mathbf{ \degreeScalar}} % Already defined by latex %\newcommand{\det}[1]{\left|#1\right|} \newcommand{\diag}[1]{\text{diag}\left(#1\right)} \newcommand{\diagonalMatrix}{\mathbf{D}} \newcommand{\diff}[2]{\frac{\text{d}#1}{\text{d}#2}} \newcommand{\diffTwo}[2]{\frac{\text{d}^2#1}{\text{d}#2^2}} \newcommand{\displacement}{x} \newcommand{\displacementVector}{\textbf{\displacement}} \newcommand{\distanceMatrix}{\mathbf{ \MakeUppercase{\distanceScalar}}} \newcommand{\distanceScalar}{d} \newcommand{\distanceVector}{\mathbf{ \distanceScalar}} \newcommand{\eigenvaltwo}{\ell} \newcommand{\eigenvaltwoMatrix}{\mathbf{L}} \newcommand{\eigenvaltwoVector}{\mathbf{l}} \newcommand{\eigenvalue}{\lambda} \newcommand{\eigenvalueMatrix}{\boldsymbol{ \Lambda}} \newcommand{\eigenvalueVector}{\boldsymbol{ \lambda}} \newcommand{\eigenvector}{\mathbf{ \eigenvectorScalar}} \newcommand{\eigenvectorMatrix}{\mathbf{U}} \newcommand{\eigenvectorScalar}{u} \newcommand{\eigenvectwo}{\mathbf{v}} \newcommand{\eigenvectwoMatrix}{\mathbf{V}} \newcommand{\eigenvectwoScalar}{v} \newcommand{\entropy}[1]{\mathcal{H}\left(#1\right)} \newcommand{\errorFunction}{E} \newcommand{\expDist}[2]{\left<#1\right>_{#2}} \newcommand{\expSamp}[1]{\left<#1\right>} \newcommand{\expectation}[1]{\left\langle #1 \right\rangle } \newcommand{\expectationDist}[2]{\left\langle #1 \right\rangle _{#2}} \newcommand{\expectedDistanceMatrix}{\mathcal{D}} \newcommand{\eye}{\mathbf{I}} \newcommand{\fantasyDim}{r} \newcommand{\fantasyMatrix}{\mathbf{ \MakeUppercase{\fantasyScalar}}} \newcommand{\fantasyScalar}{z} \newcommand{\fantasyVector}{\mathbf{ \fantasyScalar}} \newcommand{\featureStd}{\varsigma} \newcommand{\gammaCdf}[3]{\mathcal{GAMMA CDF}\left(#1|#2,#3\right)} \newcommand{\gammaDist}[3]{\mathcal{G}\left(#1|#2,#3\right)} \newcommand{\gammaSamp}[2]{\mathcal{G}\left(#1,#2\right)} \newcommand{\gaussianDist}[3]{\mathcal{N}\left(#1|#2,#3\right)} \newcommand{\gaussianSamp}[2]{\mathcal{N}\left(#1,#2\right)} \newcommand{\given}{|} \newcommand{\half}{\frac{1}{2}} \newcommand{\heaviside}{H} \newcommand{\hiddenMatrix}{\mathbf{ \MakeUppercase{\hiddenScalar}}} \newcommand{\hiddenScalar}{h} \newcommand{\hiddenVector}{\mathbf{ \hiddenScalar}} \newcommand{\identityMatrix}{\eye} \newcommand{\inducingInputScalar}{z} \newcommand{\inducingInputVector}{\mathbf{ \inducingInputScalar}} \newcommand{\inducingInputMatrix}{\mathbf{Z}} \newcommand{\inducingScalar}{u} \newcommand{\inducingVector}{\mathbf{ \inducingScalar}} \newcommand{\inducingMatrix}{\mathbf{U}} \newcommand{\inlineDiff}[2]{\text{d}#1/\text{d}#2} \newcommand{\inputDim}{q} \newcommand{\inputMatrix}{\mathbf{X}} \newcommand{\inputScalar}{x} \newcommand{\inputSpace}{\mathcal{X}} \newcommand{\inputVals}{\inputVector} \newcommand{\inputVector}{\mathbf{ \inputScalar}} \newcommand{\iterNum}{k} \newcommand{\kernel}{\kernelScalar} \newcommand{\kernelMatrix}{\mathbf{K}} \newcommand{\kernelScalar}{k} \newcommand{\kernelVector}{\mathbf{ \kernelScalar}} \newcommand{\kff}{\kernelScalar_{\mappingFunction \mappingFunction}} \newcommand{\kfu}{\kernelVector_{\mappingFunction \inducingScalar}} \newcommand{\kuf}{\kernelVector_{\inducingScalar \mappingFunction}} \newcommand{\kuu}{\kernelVector_{\inducingScalar \inducingScalar}} \newcommand{\lagrangeMultiplier}{\lambda} \newcommand{\lagrangeMultiplierMatrix}{\boldsymbol{ \Lambda}} \newcommand{\lagrangian}{L} \newcommand{\laplacianFactor}{\mathbf{ \MakeUppercase{\laplacianFactorScalar}}} \newcommand{\laplacianFactorScalar}{m} \newcommand{\laplacianFactorVector}{\mathbf{ \laplacianFactorScalar}} \newcommand{\laplacianMatrix}{\mathbf{L}} \newcommand{\laplacianScalar}{\ell} \newcommand{\laplacianVector}{\mathbf{ \ell}} \newcommand{\latentDim}{q} \newcommand{\latentDistanceMatrix}{\boldsymbol{ \Delta}} \newcommand{\latentDistanceScalar}{\delta} \newcommand{\latentDistanceVector}{\boldsymbol{ \delta}} \newcommand{\latentForce}{f} \newcommand{\latentFunction}{u} \newcommand{\latentFunctionVector}{\mathbf{ \latentFunction}} \newcommand{\latentFunctionMatrix}{\mathbf{ \MakeUppercase{\latentFunction}}} \newcommand{\latentIndex}{j} \newcommand{\latentScalar}{z} \newcommand{\latentVector}{\mathbf{ \latentScalar}} \newcommand{\latentMatrix}{\mathbf{Z}} \newcommand{\learnRate}{\eta} \newcommand{\lengthScale}{\ell} \newcommand{\rbfWidth}{\ell} \newcommand{\likelihoodBound}{\mathcal{L}} \newcommand{\likelihoodFunction}{L} \newcommand{\locationScalar}{\mu} \newcommand{\locationVector}{\boldsymbol{ \locationScalar}} \newcommand{\locationMatrix}{\mathbf{M}} \newcommand{\variance}[1]{\text{var}\left( #1 \right)} \newcommand{\mappingFunction}{f} \newcommand{\mappingFunctionMatrix}{\mathbf{F}} \newcommand{\mappingFunctionTwo}{g} \newcommand{\mappingFunctionTwoMatrix}{\mathbf{G}} \newcommand{\mappingFunctionTwoVector}{\mathbf{ \mappingFunctionTwo}} \newcommand{\mappingFunctionVector}{\mathbf{ \mappingFunction}} \newcommand{\scaleScalar}{s} \newcommand{\mappingScalar}{w} \newcommand{\mappingVector}{\mathbf{ \mappingScalar}} \newcommand{\mappingMatrix}{\mathbf{W}} \newcommand{\mappingScalarTwo}{v} \newcommand{\mappingVectorTwo}{\mathbf{ \mappingScalarTwo}} \newcommand{\mappingMatrixTwo}{\mathbf{V}} \newcommand{\maxIters}{K} \newcommand{\meanMatrix}{\mathbf{M}} \newcommand{\meanScalar}{\mu} \newcommand{\meanTwoMatrix}{\mathbf{M}} \newcommand{\meanTwoScalar}{m} \newcommand{\meanTwoVector}{\mathbf{ \meanTwoScalar}} \newcommand{\meanVector}{\boldsymbol{ \meanScalar}} \newcommand{\mrnaConcentration}{m} \newcommand{\naturalFrequency}{\omega} \newcommand{\neighborhood}[1]{\mathcal{N}\left( #1 \right)} \newcommand{\neilurl}{http://inverseprobability.com/} \newcommand{\noiseMatrix}{\boldsymbol{ E}} \newcommand{\noiseScalar}{\epsilon} \newcommand{\noiseVector}{\boldsymbol{ \epsilon}} \newcommand{\norm}[1]{\left\Vert #1 \right\Vert} \newcommand{\normalizedLaplacianMatrix}{\hat{\mathbf{L}}} \newcommand{\normalizedLaplacianScalar}{\hat{\ell}} \newcommand{\normalizedLaplacianVector}{\hat{\mathbf{ \ell}}} \newcommand{\numActive}{m} \newcommand{\numBasisFunc}{m} \newcommand{\numComponents}{m} \newcommand{\numComps}{K} \newcommand{\numData}{n} \newcommand{\numFeatures}{K} \newcommand{\numHidden}{h} \newcommand{\numInducing}{m} \newcommand{\numLayers}{\ell} \newcommand{\numNeighbors}{K} \newcommand{\numSequences}{s} \newcommand{\numSuccess}{s} \newcommand{\numTasks}{m} \newcommand{\numTime}{T} \newcommand{\numTrials}{S} \newcommand{\outputIndex}{j} \newcommand{\paramVector}{\boldsymbol{ \theta}} \newcommand{\parameterMatrix}{\boldsymbol{ \Theta}} \newcommand{\parameterScalar}{\theta} \newcommand{\parameterVector}{\boldsymbol{ \parameterScalar}} \newcommand{\partDiff}[2]{\frac{\partial#1}{\partial#2}} \newcommand{\precisionScalar}{j} \newcommand{\precisionVector}{\mathbf{ \precisionScalar}} \newcommand{\precisionMatrix}{\mathbf{J}} \newcommand{\pseudotargetScalar}{\widetilde{y}} \newcommand{\pseudotargetVector}{\mathbf{ \pseudotargetScalar}} \newcommand{\pseudotargetMatrix}{\mathbf{ \widetilde{Y}}} \newcommand{\rank}[1]{\text{rank}\left(#1\right)} \newcommand{\rayleighDist}[2]{\mathcal{R}\left(#1|#2\right)} \newcommand{\rayleighSamp}[1]{\mathcal{R}\left(#1\right)} \newcommand{\responsibility}{r} \newcommand{\rotationScalar}{r} \newcommand{\rotationVector}{\mathbf{ \rotationScalar}} \newcommand{\rotationMatrix}{\mathbf{R}} \newcommand{\sampleCovScalar}{s} \newcommand{\sampleCovVector}{\mathbf{ \sampleCovScalar}} \newcommand{\sampleCovMatrix}{\mathbf{s}} \newcommand{\scalarProduct}[2]{\left\langle{#1},{#2}\right\rangle} \newcommand{\sign}[1]{\text{sign}\left(#1\right)} \newcommand{\sigmoid}[1]{\sigma\left(#1\right)} \newcommand{\singularvalue}{\ell} \newcommand{\singularvalueMatrix}{\mathbf{L}} \newcommand{\singularvalueVector}{\mathbf{l}} \newcommand{\sorth}{\mathbf{u}} \newcommand{\spar}{\lambda} \newcommand{\trace}[1]{\text{tr}\left(#1\right)} \newcommand{\BasalRate}{B} \newcommand{\DampingCoefficient}{C} \newcommand{\DecayRate}{D} \newcommand{\Displacement}{X} \newcommand{\LatentForce}{F} \newcommand{\Mass}{M} \newcommand{\Sensitivity}{S} \newcommand{\basalRate}{b} \newcommand{\dampingCoefficient}{c} \newcommand{\mass}{m} \newcommand{\sensitivity}{s} \newcommand{\springScalar}{\kappa} \newcommand{\springVector}{\boldsymbol{ \kappa}} \newcommand{\springMatrix}{\boldsymbol{ \mathcal{K}}} \newcommand{\tfConcentration}{p} \newcommand{\tfDecayRate}{\delta} \newcommand{\tfMrnaConcentration}{f} \newcommand{\tfVector}{\mathbf{ \tfConcentration}} \newcommand{\velocity}{v} \newcommand{\sufficientStatsScalar}{g} \newcommand{\sufficientStatsVector}{\mathbf{ \sufficientStatsScalar}} \newcommand{\sufficientStatsMatrix}{\mathbf{G}} \newcommand{\switchScalar}{s} \newcommand{\switchVector}{\mathbf{ \switchScalar}} \newcommand{\switchMatrix}{\mathbf{S}} \newcommand{\tr}[1]{\text{tr}\left(#1\right)} \newcommand{\loneNorm}[1]{\left\Vert #1 \right\Vert_1} \newcommand{\ltwoNorm}[1]{\left\Vert #1 \right\Vert_2} \newcommand{\onenorm}[1]{\left\vert#1\right\vert_1} \newcommand{\twonorm}[1]{\left\Vert #1 \right\Vert} \newcommand{\vScalar}{v} \newcommand{\vVector}{\mathbf{v}} \newcommand{\vMatrix}{\mathbf{V}} \newcommand{\varianceDist}[2]{\text{var}_{#2}\left( #1 \right)} % Already defined by latex %\newcommand{\vec}{#1:} \newcommand{\vecb}[1]{\left(#1\right):} \newcommand{\weightScalar}{w} \newcommand{\weightVector}{\mathbf{ \weightScalar}} \newcommand{\weightMatrix}{\mathbf{W}} \newcommand{\weightedAdjacencyMatrix}{\mathbf{A}} \newcommand{\weightedAdjacencyScalar}{a} \newcommand{\weightedAdjacencyVector}{\mathbf{ \weightedAdjacencyScalar}} \newcommand{\onesVector}{\mathbf{1}} \newcommand{\zerosVector}{\mathbf{0}} $$

The Promise of AI

  • Automation forces humans to adapt, we serve.

  • We can only automate by systemizing and controlling environment.

  • AI promises to be first wave of automation that adapts to us rather than us to it.

That Promise …

… will remain unfulfilled with current systems design.

Artificial vs Natural Systems [edit]

Let’s take a step back from artificial intelligence, and consider natural intelligence. Or even more generally, let’s consider the contrast between an artificial system and an natural system. The key difference between the two is that artificial systems are designed whereas natural systems are evolved.

Systems design is a major component of all Engineering disciplines. The details differ, but there is a single common theme: achieve your objective with the minimal use of resources to do the job. That provides efficiency. The engineering designer imagines a solution that requires the minimal set of components to achieve the result. A water pump has one route through the pump. That minimises the number of components needed. Redundancy is introduced only in safety critical systems, such as aircraft control systems. Students of biology, however, will be aware that in nature system-redundancy is everywhere. Redundancy leads to robustness. For an organism to survive in an evolving environment it must first be robust, then it can consider how to be efficient. Indeed, organisms that evolve to be too efficient at a particular task, like those that occupy a niche environment, are particularly vulnerable to extinction.

This notion is akin to the idea that only the best will survive, popularly encoded into an notion of evolution by Herbert Spencer’s quote.

Survival of the fittest

Herbet Spencer, 1864

Darwin himself never said “Survival of the Fittest” he talked about evolution by natural selection.

Non-survival of the non-fit

Evolution is better described as “non-survival of the non-fit”. You don’t have to be the fittest to survive, you just need to avoid the pitfalls of life. This is the first priority.

So it is with natural vs artificial intelligences. Any natural intelligence that was not robust to changes in its external environment would not survive, and therefore not reproduce. In contrast the artificial intelligences we produce are designed to be efficient at one specific task: control, computation, playing chess. They are fragile.

The first rule of a natural system is not be intelligent, it is “don’t be stupid”.

A mistake we make in the design of our systems is to equate fitness with the objective function, and to assume it is known and static. In practice, a real environment would have an evolving fitness function which would be unknown at any given time.

You can also check my blog post on “Natural and Artificial Intelligence”.

The first criterion of a natural intelligence is don’t fail, not because it has a will or intent of its own, but because if it had failed it wouldn’t have stood the test of time. It would no longer exist. In contrast, the mantra for artificial systems is to be more efficient. Our artificial systems are often given a single objective (in machine learning it is encoded in a mathematical function) and they aim to achieve that objective efficiently. These are different characteristics. Even if we wanted to incorporate don’t fail in some form, it is difficult to design for. To design for “don’t fail”, you have to consider every which way in which things can go wrong, if you miss one you fail. These cases are sometimes called corner cases. But in a real, uncontrolled environment, almost everything is a corner. It is difficult to imagine everything that can happen. This is why most of our automated systems operate in controlled environments, for example in a factory, or on a set of rails. Deploying automated systems in an uncontrolled environment requires a different approach to systems design. One that accounts for uncertainty in the environment and is robust to unforeseen circumstances.

The systems we produce today only work well when their tasks are pigeonholed, bounded in some way. To achieve robust artificial intelligences we need new approaches to both the design of the individual components, and the combination of components within our AI systems. We need to deal with uncertainty and increase robustness. Today, it is easy to make a fool of an artificial intelligent agent, technology needs to address the challenge of the uncertain environment to achieve robust intelligences.

However, even if we find technological solutions for these challenges, it may be that the essence of human intelligence remains out of reach. It may be that the most quintessential element of our intelligence is defined by limitations. Limitations that computers have never experienced.

Claude Shannon developed the idea of information theory: the mathematics of information. He defined the amount of information we gain when we learn the result of a coin toss as a “bit” of information. A typical computer can communicate with another computer with a billion bits of information per second. Equivalent to a billion coin tosses per second. So how does this compare to us? Well, we can also estimate the amount of information in the English language. Shannon estimated that the average English word contains around 12 bits of information, twelve coin tosses, this means our verbal communication rates are only around the order of tens to hundreds of bits per second. Computers communicate tens of millions of times faster than us, in relative terms we are constrained to a bit of pocket money, while computers are corporate billionaires.

Our intelligence is not an island, it interacts, it infers the goals or intent of others, it predicts our own actions and how we will respond to others. We are social animals, and together we form a communal intelligence that characterises our species. For intelligence to be communal, our ideas to be shared somehow. We need to overcome this bandwidth limitation. The ability to share and collaborate, despite such constrained ability to communicate, characterises us. We must intellectually commune with one another. We cannot communicate all of what we saw, or the details of how we are about to react. Instead, we need a shared understanding. One that allows us to infer each other’s intent through context and a common sense of humanity. This characteristic is so strong that we anthropomorphise any object with which we interact. We apply moods to our cars, our cats, our environment. We seed the weather, volcanoes, trees with intent. Our desire to communicate renders us intellectually animist.

But our limited bandwidth doesn’t constrain us in our imaginations. Our consciousness, our sense of self, allows us to play out different scenarios. To internally observe how our self interacts with others. To learn from an internal simulation of the wider world. Empathy allows us to understand others’ likely responses without having the full detail of their mental state. We can infer their perspective. Self-awareness also allows us to understand our own likely future responses, to look forward in time, play out a scenario. Our brains contain a sense of self and a sense of others. Because our communication cannot be complete it is both contextual and cultural. When driving a car in the UK a flash of the lights at a junction concedes the right of way and invites another road user to proceed, whereas in Italy, the same flash asserts the right of way and warns another road user to remain.

Our main intelligence is our social intelligence, intelligence that is dedicated to overcoming our bandwidth limitation. We are individually complex, but as a society we rely on shared behaviours and oversimplification of our selves to remain coherent.

This nugget of our intelligence seems impossible for a computer to recreate directly, because it is a consequence of our evolutionary history. The computer, on the other hand, was born into a world of data, of high bandwidth communication. It was not there through the genesis of our minds and the cognitive compromises we made are lost to time. To be a truly human intelligence you need to have shared that journey with us.

Of course, none of this prevents us emulating those aspects of human intelligence that we observe in humans. We can form those emulations based on data. But even if an artificial intelligence can emulate humans to a high degree of accuracy it is a different type of intelligence. It is not constrained in the way human intelligence is. You may ask does it matter? Well, it is certainly important to us in many domains that there’s a human pulling the strings. Even in pure commerce it matters: the narrative story behind a product is often as important as the product itself. Handmade goods attract a price premium over factory made. Or alternatively in entertainment: people pay more to go to a live concert than for streaming music over the internet. People will also pay more to go to see a play in the theatre rather than a movie in the cinema.

In many respects I object to the use of the term Artificial Intelligence. It is poorly defined and means different things to different people. But there is one way in which the term is very accurate. The term artificial is appropriate in the same way we can describe a plastic plant as an artificial plant. It is often difficult to pick out from afar whether a plant is artificial or not. A plastic plant can fulfil many of the functions of a natural plant, and plastic plants are more convenient. But they can never replace natural plants.

In the same way, our natural intelligence is an evolved thing of beauty, a consequence of our limitations. Limitations which don’t apply to artificial intelligences and can only be emulated through artificial means. Our natural intelligence, just like our natural landscapes, should be treasured and can never be fully replaced.

Technical Consequence

  • The gap between the game and reality.

  • The need for extrapolation over interpolation.

Figure: Some software components in a ride allocation system. Circled components are hypothetical, rectangles represent actual data.

Machine Learning Systems Design [edit]

The challenges of integrating different machine learning components into a whole that acts effectively as a system seem unresolved. In software engineering, separating parts of a system in this way is known as component-based software engineering. The core idea is that the different parts of the system can be independently designed according to a sub-specfication. This is sometimes known as separation of concerns. However, once the components are machine learning based, tighter coupling becomes a side effect of the learned nature of the system. For example if a driverless car’s detection of cyclist is dependent on its detection of the road surface, a change in the road surface detection algorithm will have downstream effects on the cyclist detection. Even if the road detection system has been improved by objective measures, the cyclist detection system may have become sensitive to the foibles of the previous version of road detection and will need to be retrained.

Most of our experience with deployment relies on some approximation to the component based model, this is also important for verification of the system. If the components of the system can be verified then the composed system can also, potentially, be verified.

The Data Crisis [edit]

Anecdotally, talking to data modelling scientists. Most say they spend 80% of their time acquiring and cleaning data. This is precipitating what I refer to as the “data crisis”. This is an analogy with software. The “software crisis” was the phenomenon of inability to deliver software solutions due to increasing complexity of implementation. There was no single shot solution for the software crisis, it involved better practice (scrum, test orientated development, sprints, code review), improved programming paradigms (object orientated, functional) and better tools (CVS, then SVN, then git).

However, these challenges aren’t new, they are merely taking a different form. From the computer’s perspective software is data. The first wave of the data crisis was known as the software crisis.

The Software Crisis

In the late sixties early software programmers made note of the increasing costs of software development and termed the challenges associated with it as the “Software Crisis”. Edsger Dijkstra referred to the crisis in his 1972 Turing Award winner’s address.

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 (1930-2002), The Humble Programmer

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. What we need is cheap high-quality data. That implies that we develop processes for improving and verifying data quality that are efficient.

There would seem to be two ways for improving efficiency. Firstly, we should not duplicate work. Secondly, where possible we should automate work.

What I term “The Data Crisis” is the modern equivalent of this problem. The quantity of modern data, and the lack of attention paid to data as it is initially “laid down” and the costs of data cleaning are bringing about a crisis in data-driven decision making. This crisis is at the core of the challenge of technical debt in machine learning (Sculley et al. 2015).

Just as with software, the crisis is most correctly addressed by ‘scaling’ the manner in which we process our data. Duplication of work occurs because the value of data cleaning is not correctly recognised in management decision making processes. Automation of work is increasingly possible through techniques in “artificial intelligence”, but this will also require better management of the data science pipeline so that data about data science (meta-data science) can be correctly assimilated and processed. The Alan Turing institute has a program focussed on this area, AI for Data Analytics.

Computer Science Paradigm Shift

  • Turing machine:
    • Code and data integrated
  • Today:
    • Code and data separated for security
  • Machine learning:
    • Software is data
  • Machine learning is a high level breach of the software/data separation.

Peppercorns [edit]

Figure: A peppercorn is a system design failure which is not a bug, but a conformance to design specification that causes problems when the system is deployed in the real world with mischevious and adversarial actors.

Asking Siri “What is a trillion to the power of a thousand minus one?” leads to a 30 minute response1 consisting of only 9s. I found this out because my nine year old grabbed my phone and did it. The only way to stop Siri was to force closure. This is an interesting example of a system feature that’s not a bug, in fact it requires clever processing from Wolfram Alpha. But it’s an unexpected result from the system performing correctly.

This challenge of facing a circumstance that was unenvisaged in design but has consequences in deployment becomes far larger when the environment is uncontrolled. Or in the extreme case, where actions of the intelligent system effect the wider environment and change it.

These unforseen circumstances are likely to lead to need for much more efficient turn-around and update for our intelligent systems. Whether we are correcting for security flaws (which are bugs) or unenvisaged circumstantial challenges: an issue I’m referring to as peppercorns. Rapid deployment of system updates is required. For example, Apple have “fixed” the problem of Siri returning long numbers.

The challenge is particularly acute because of the scale at which we can deploy AI solutions. This means when something does go wrong, it may be going wrong in billions of households simultaneously.

You can also check my blog post on “Decision Making and Diversity”. You can also check my blog post on “Natural vs Artifical Intelligence”.

Emulation [edit]

Figure: Real world systems consiste of simulators, that capture our domain knowledge about how our systems operate. Different simulators run at different speeds and granularities.

In many real world systems, decisions are made through simulating the environment. Simulations may operate at different granularities. For example, simulations are used in weather forecasts and climate forecasts. The UK Met office uses the same code for both, but operates climate simulations one at greater spatial and temporal resolutions.

Figure: A statistical emulator is a system that reconstructs the simulation with a statistical model.

A statistical emulator is a data-driven model that learns about the underlying simulation. Importantly, learns with uncertainty, so it ‘knows what it doesn’t know’. In practice, we can call the emulator in place of the simulator. If the emulator ‘doesn’t know’, it can call the simulator for the answer.

Figure: A statistical emulator is a system that reconstructs the simulation with a statistical model. As well as reconstructing the simulation, a statistical emulator can be used to correlate with the real world.

Figure: In modern machine learning system design, the emulator may also consider the output of ML models (for monitoring bias or accuracy) and Operations Research models..

As well as reconstructing an individual simulator, the emulator can calibrate the simulation to the real world, by monitoring differences between the simulator and real data. This allows the emulator to characterise where the simulation can be relied on, i.e. we can validate the simulator.

Similarly, the emulator can adjudicate between simulations. This is known as multi-fidelity emulation. The emulator characterizes which emulations perform well where.

If all this modelling is done with judiscious handling of the uncertainty, the computational doubt, then the emulator can assist in desciding what experiment should be run next to aid a decision: should we run a simulator, in which case which one, or should we attempt to acquire data from a real world intervention.

Bayesian Inference by Rejection Sampling [edit]

One view of Bayesian inference is to assume we are given a mechanism for generating samples, where we assume that mechanism is representing on accurate view on the way we believe the world works.

This mechanism is known as our prior belief.

We combine our prior belief with our observations of the real world by discarding all those samples that are inconsistent with our prior. The likelihood defines mathematically what we mean by inconsistent with the prior. The higher the noise level in the likelihood, the looser the notion of consistent.

The samples that remain are considered to be samples from the posterior.

This approach to Bayesian inference is closely related to two sampling techniques known as rejection sampling and importance sampling. It is realized in practice in an approach known as approximate Bayesian computation (ABC) or likelihood-free inference.

In practice, the algorithm is often too slow to be practical, because most samples will be inconsistent with the data and as a result the mechanism has to be operated many times to obtain a few posterior samples.

However, in the Gaussian process case, when the likelihood also assumes Gaussian noise, we can operate this mechanism mathematically, and obtain the posterior density analytically. This is the benefit of Gaussian processes.

Figure: One view of Bayesian inference is we have a machine for generating samples (the prior), and we discard all samples inconsistent with our data, leaving the samples of interest (the posterior). The Gaussian process allows us to do this analytically.

Deep Emulation [edit]

Figure: A potential path of models in a machine learning system.

As a solution we can use of emulators. When constructing an ML system, software engineers, ML engineers, economists and operations researchers are explicitly defining relationships between variables of interest in the system. That implicitly defines a joint distribution, $p(\dataVector^*, \dataVector)$. In a decomposable system any sub-component may be defined as $p(\dataVector_\mathbf{i}|\dataVector_\mathbf{j})$ where $\dataVector_\mathbf{i}$ and $\dataVector_\mathbf{j}$ represent sub-sets of the full set of variables $\left\{\dataVector^*, \dataVector \right\}$. In those cases where the relationship is deterministic, the probability density would collapse to a vector-valued deterministic function, $\mappingFunctionVector_\mathbf{i}\left(\dataVector_\mathbf{j}\right)$.

Inter-variable relationships could be defined by, for example a neural network (machine learning), an integer program (operational research), or a simulation (supply chain). This makes probabilistic inference in this joint density for real world systems is either very hard or impossible.

Emulation is a form of meta-modelling: we construct a model of the model. We can define the joint density of an emulator as $s(\dataVector*, \dataVector)$, but if this probability density is to be an accurate representation of our system, it is likely to be prohibitively complex. Current practice is to design an emulator to deal with a specific question. This is done by fitting an ML model to a simulation from the the appropriate conditional distribution, $p(\dataVector_\mathbf{i}|\dataVector_\mathbf{j})$, which is intractable. The emulator provides an approximated answer of the form $s(\dataVector_\mathbf{i}|\dataVector_\mathbf{j})$. Critically, an emulator should incorporate its uncertainty about its approximation. So the emulator answer will be less certain than direct access to the conditional $p(\dataVector_i|\dataVector_j)$, but it may be sufficiently confident to act upon. Careful design of emulators to answer a given question leads to efficient diagnostics and understanding of the system. But in a complex interacting system an exponentially increasing number of questions can be asked. This calls for a system of automated construction of emulators which selects the right structure and redeploys the emulator as necessary. Rapid redeployment of emulators could exploit pre-existing emulators through transfer learning.

Automatically deploying these families of emulators for full system understanding is highly ambitious. It requires advances in engineering infrastructure, emulation and Bayesian optimization. However, the intermediate steps of developing this architecture also allow for automated monitoring of system accuracy and fairness. This facilitates AutoML on a component-wise basis which we can see as a simple implementation of AutoAI. The proposal is structured so that despite its technical ambition there is a smooth ramp of benefits to be derived across the programme of work.

In Applied Mathematics, the field studying these techniques is known as uncertainty quantification. The new challenge is the automation of emulator creation on demand to answer questions of interest and facilitate the system design, i.e. AutoAI through BSO.

At design stage, any particular AI task could be decomposed in multiple ways. Bayesian system optimization will assist both in determining the large-scale system design through exploring different decompositions and in refinement of the deployed system.

So far, most work on emulators has focussed on emulating a single component. Automated deployment and maintenance of ML systems requires networks of emulators that can be deployed and redeployed on demand depending on the particular question of interest. Therefore, the technical innovations we require are in the mathematical composition of emulator models (Damianou and Lawrence 2013; Perdikaris et al. 2017). Different chains of emulators will need to be rapidly composed to make predictions of downstream performance. This requires rapid retraining of emulators and propagation of uncertainty through the emulation pipeline a process we call deep emulation.

Recomposing the ML system requires structural learning of the network. By parameterizing covariance functions appropriately this can be done through Gaussian processes (e.g. (Damianou et al., n.d.)), but one could also consider Bayesian neural networks and other generative models, e.g. Generative Adversarial Networks (Goodfellow et al. 2014).

Figure: A potential path of models in a machine learning system.

Figure: A potential path of models in a machine learning system.

Figure: A statistical emulator is a system that reconstructs the simulation with a statistical model. As well as reconstructing the simulation, a statistical emulator can be used to correlate with the real world.

Conclusion

  • AI is fundamentally ML System Design
  • We are not ready to deploy automation in uncontrolled environments.
  • Until we can monitoring and update will be key.

References

Damianou, Andreas, Carl Henrik Ek, Michalis K. Titsias, and Neil D. Lawrence. n.d. “Manifold Relevance Determination.” In.

Damianou, Andreas, and Neil D. Lawrence. 2013. “Deep Gaussian Processes.” In, 31:207–15.

Goodfellow, Ian, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. 2014. “Generative Adversarial Nets.” In Advances in Neural Information Processing Systems 27, edited by Z. Ghahramani, M. Welling, C. Cortes, N. D. Lawrence, and K. Q. Weinberger, 2672–80. Curran Associates, Inc.

Perdikaris, Paris, Maziar Raissi, Andreas Damianou, Neil D. Lawrence, and George Em Karnidakis. 2017. “Nonlinear Information Fusion Algorithms for Data-Efficient Multi-Fidelity Modelling.” Proc. R. Soc. A 473 (20160751). https://doi.org/10.1098/rspa.2016.0751.

Sculley, D., Gary Holt, Daniel Golovin, Eugene Davydov, Todd Phillips, Dietmar Ebner, Vinay Chaudhary, Michael Young, Jean-François Crespo, and Dan Dennison. 2015. “Hidden Technical Debt in Machine Learning Systems.” In Advances in Neural Information Processing Systems 28, edited by Corinna Cortes, Neil D. Lawrence, Daniel D. Lee, Masashi Sugiyama, and Roman Garnett, 2503–11. Curran Associates, Inc. http://papers.nips.cc/paper/5656-hidden-technical-debt-in-machine-learning-systems.pdf.


  1. Apple has fixed this issue so that Siri no longer does this.