Europe Americas Asia-Pacific International

  • AI in Equipment and Auto Finance

    Part 2: Using Machine Learning in the Wild

  • REQUEST A DEMO

  • esg

  • Investors

  • Contact

    Close
Company
Product
Careers

Company
Product
Careers
Perspectives

AI in Equipment and Auto Finance

Part 2: Using Machine Learning in the Wild

This is our second paper on artificial intelligence in the asset finance industry. Part 2: Using Machine Learning in the Wild is a more technical follow-up to 2019's Balancing Risk and Reward, exploring in detail two specific use cases which take very different approaches to machine learning implementation.

In this paper, we show how auto and equipment finance providers can leverage machine learning and artificial intelligence techniques to enhance efficiency and improve customer service. With worked-through examples, Part 2: Using Machine Learning in the Wild will help finance providers decide which approach would be best for their business and how others in the industry are currently engaging with this new field of technology.

See also Part 1: Balancing Risk and Reward and Part 3: Moving Forward with Machine Learning.

Download Now
  • AI in Asset Finance

    Over the last decade, machine learning techniques for object recognition have progressed at an impressive rate. This has made the concept of self-driving cars not just possible but achievable, which has fuelled further research into the optimisation of ML techniques in this area.

  • AI-as-a-Service

    Alongside advancements in the way ML models are trained, cloud providers have taken the opportunity to automate the surrounding infrastructure. AIaaS products reduce the time needed to set up the hardware and software required for an experiment or proof-of-concept exercise.

  • Uses of AI at Alfa

    At Alfa, we automate the recording of end-to-end test failures along with its root cause and resulting fix. This data is stored efficiently and captures many years worth of failures. This is useful for training a neural network to recognise patterns in failures.

The two use cases outlined in this paper demonstrate two very different approaches to using ML to solve a problem: one that relies on AI-as-a-Service, and one that creates an in-house framework for developing ML models.

The use cases

 

Automated object recognition for vehicle licence plates

 

This use case draws on AIaaS products to integrate licence plate recognition with business logic in Alfa’s own asset finance software platform, Alfa Systems. It demonstrates how these services enable experimentation with ML techniques, at little cost and without requiring great expertise.

 

End-to-end test analyser

 

Picking up from our first paper on AI, this use case analyses the results of the automated testing of Alfa's internal source code - this time using our own reusable framework. A neural network trained on historic failures learns patterns between the areas of code that are causing test failures, and the failures themselves.

  • Neural Networks

    A machine learning algorithm modelled on the way our brains’ neurons interact with each other. Neural networks commonly have layers of these neurons: an input layer, an output layer, and one or more middle layers (also known as "hidden" layers).

  • Deep Neural Network

    A neural network that has more than one "hidden" layer.

  • Convolutional Neural Network

    A neural network that uses a convolutional operation to determine the activation of neurons in the next layer - rather than mathematical multiplication of the synapse weights and current layer's neuron values.

  • Residual Neural Network

    A deep neural network that uses techniques during the training process that allow it to retain accuracy when trained over very large datasets and with a high number of hidden layers.

Which approach is right for you?

Flow Chart20210119 Updated

The experiments detailed in this paper explore different ways to develop ML solutions, and each requires different levels of time, effort and expertise.

However, both solutions rely on domain knowledge of the data used; the first relies heavily on research in object recognition and a resulting dataset that was created by domain experts in that field, while the second is based on data with which most people at Alfa are familiar.

This paper features a foreword from Blaise Thomson, whose speech technology start-up VocalIQ was acquired by Apple and formed an important part of the Siri development team.


AlfaIQ_Logo_Sm_200908_091537.png?mtime=20200908091537#asset:3168

Coming Soon... Alfa iQ The intelligence platform for asset finance. Learn more and stay updated at alfaiq.com.

  • Fp Layout
    Document

    AI in Equipment and Auto Finance Part 2: Using Machine Learning in the Wild

    Download Now 19/09/2021