Highlights

 

New MPAI-CUI V2.0 Call for Technologies presented on 8 January

 

They Compression and Understand of Industrial Data (MPAI-CUI) V2.0 will be presented online according to the schedule reported below.

 

Type Register for presentation of Code Date/Time (UTC)
Call for Techologies Compression and Understanding of Industrial Data V2.0 CUI 2025/01/08T15

 

The Company Performance Prediction Use Case of the MPAI-CUI V2.0

Compression and Understanding of Industrial Data (MPAI-CUI) was one of the first standards produced by MPAI. It supported the Company Performance Prediction (CUI-CPP) Use Case that was based on the notion that the future of a company largely depends on how it is structured, its financial state, and the risks it may face in the future. The complex task of creating a standard to predict the future of a company using such variables was successfully achieved. The solution used a set of descriptors into which Governance Data and Financial Data should be converted, passing the descriptors to a neural network, and handling the company-assessed Risks with parameters.

Governance Descriptors were: #Stakeholder Individuals, #Stakeholder Companies, Shareholder Share, Shareholders Gender, Decision-Makers Gender, #Decision-Makers, Members of the Revision And Advisory Board, Presence Of The Advisory Company, #Decision-Makers By The Same Family, Company Phase (Age).

Financial Descriptors were: Revenues, EBITDA Margin, EBITDA, Quick Ratio, Current Ratio, Net Working Capital, Net Financial Position, Net Short-Term Assets, Shareholder Funds-Fixed Assets, Long-Term Liability Ratio, Coverage Of Fixed Assets, Amortisation Rate, Debt On Sales

Interest Coverage Ratio, Average Stock Turnover, Stock Coverage Days, Return On Investments (ROI), Return On Assets (ROA), Return On Sales (ROS), Return On Equity (ROE), Cash Flow, Interest On Sales, Type Of Financial Statement.

The company being examined provided their assessment of the risks which were converted to a Risk Matrix included the following characteristics: Occurrence (3 values), Business Impact (3 values), Gravity (5 values), Risk retention (portion of the risk that the Company decides to retain).

The Governance and Financial Descriptors together with the Prediction Horizon fed to a properly trained Discontinuity and Default Prediction Neural Network provided an Organisational Model Index and a Default Probability. Default Probability and Risk Matrix fed to a Prediction Result Perturbation AIM which perturbed the Default Probability and produced the Business Discontinuity Probability.

Figure 1 depicts the AI Workflow that performs as described above.

Figure 1 – The Company Performance Prediction AI Workflow of MPAI-CUI V1.1

CUI-CPP V2.0 is substantially more ambitious because it targets

  • A more precise identification of Cyber, Digitisation, Climate, and Business risks.
  • A definition of risks:
    • Risk name, Risk type: (cyber, etc.),
    • Target regulation,
    • Vector of inputs which could include, e.g. for Cyber Risks:
      • Name of input: IP address, Denial of service;
      • Time: time the attack was detected;
      • Source: provider of input vector; Type: image, text, category, etc.;
      • Value: depends on type.

With this additional information, it is possible to train a neural network – or a set of neural networks – that receive(s) Risk Data in addition to Prediction Horizon, and Governance and Financial Descriptors, The index and probability outputs of CUI-CPP V1.1 now become Descriptors, data structures that still include index or probability but also information on which input elements have more influence on index and probabilities.

Figure 2 – The Company Performance Prediction AI Workflow of MPAI-CUI V1.1

Figure 2 retains Risk Assessment and Risk Matrix used when sufficient data are not available to train the Neural Network or when the Neural Network may not be used because it does not comply with relevant regulations.

At the 51st MPAI General Assembly (MPAI-51), MPAI published a Call for MPAI-CUI V2.0 Technologies requesting proposals for all the Data Types and the AIMs in Figure 2. All parties having rights to technologies satisfying the Use Cases and Functional Requirements and the Framework Licence of the planned Technical Specification MPAI-CUI V2.0 are invited to respond to the Call for Technologies, preferably using the Template for Responses. Submissions received by 2024/02/11 will be assessed and considered for use in the development of said MPAI-CUI Technical Specification.

 

Reusability of Data Types in MPAI

One of the foundations of MPAI is “reusability”. As far as possible, a standardisation item – an AI Module (AIM) or a Data Type – should be specified in such a way that it can be reused in as many standards as possible.

Human and Machine Communication (MPAI-HMC) provides a good example of how this basic guideline has been implemented: MPAI-HMC only specifies two AI Modules. All the other AIMs are borrowed from four other standards: Context-based Audio Enhancement (MPAI-CAE), Multimodal Conversation (MPAI-MMC), Object and Scene Description (MPAI-OSD), and Portable Avatar Format (MPAI-PAF).

The table below lists all the AIMs used by MPAI-HMC V2.0. AIMs in bold are Composite AIMs.

 

CAE MMC OSD PAF
Audio Analysis Transform Automatic Speech Recognition Audio-Visual Alignment Audio-Visual Scene Rendering
Audio Descriptors Multiplexing Entity Dialogue Processing Audio-Visual Event Description Face Identity Recognition
Audio Object Identification Entity Speech Description Audio-Visual Scene Demultiplexing Entity Body Description
Audio Scene Description Entity Text Description Audio-Visual Scene Description Entity Face Description
Audio Separation and Enhancement Natural Language Understanding Visual Direction Identification Portable Avatar Demultiplexing
Audio Source Localisation Personal Status Extraction Visual Instance Identification PS-Face Interpretation
Audio Synthesis Transform Personal Status Multiplexing Visual Object Extraction PS-Gesture Interpretation
HMC PS-Speech Interpretation Visual Object Identification Personal Status Display
AV Scene Integration and Description PS-Text Interpretation Visual Scene Description Portable Avatar Multiplexing
Entity and Context Understanding Speaker Identity Recognition
Text and Speech Translation
Text-To-Speech
Text-to-Text Translation

 

A similar table can be built with all data types used – and none of which is specified – by MPAI-HMC V2.0.

 

MPAI-AIF MPAI-MMC MPAI-OSD MPAI-PAF
Machine Learning Model Cognitive State Audio-Visual Basic Scene Descriptors 3D Model
MPAI-CAE Emotion Audio-Visual Basic Scene Geometry Avatar
Audio Basic Scene Descriptors Intention Audio-Visual Event Descriptors Body Descriptors
Audio Basic Scene Geometry Meaning Audio-Visual Object Face Descriptors
Audio Object Personal Status Audio-Visual Scene Descriptors Portable Avatar
Audio Scene Descriptors Social Attitude Audio-Visual Scene Geometry
Audio Scene Geometry Speech Descriptors Instance Identifier
Speech Object Point of View
Text Descriptors Selector
Text Object Space-Time
Spatial Attitude
Time
Visual Basic Scene Descriptors
Visual Basic Scene Geometry
Visual Object
Visual Scene Descriptors
Visual Scene Geometry

 

 

It is easy to conclude that, more than a philosophy, AIM and Data Type reusability is a necessity.

 

Meetings in the coming January meeting cycle

Group name 30 Dec -03 Jan 06-10 Jan 13-17 Jan 20-24 Jan Time (UTC)
AI Framework 20 16
AI-based End-to-End Video Coding 8 15 14
AI-Enhanced Video Coding 8 21 13
8 14
Artificial Intelligence for Health 16 10
Communication 2 16 13:30
Compression & Understanding of Industrial Data 8(*)  15 15
Connected Autonomous Vehicle 2 9 16 23 16
Context-based Audio enhancement 7 14 21 17
Industry and Standards 3 17 16
MPAI Metaverse Model 3 10 17 24 15
Multimodal Conversation 7 14 21 14
Neural Network Watermarking 7 14 21 15
Portable Avatar Format 3 10 17 24 11
Server-based Pred. Multipl. Gaming 2 9 16 23 15
XR – Live theatrical Performance 31 7 14 21 18
XRV – Collab.  immersive laboratory 2 9 16 17
General Assembly (MPAI-52)       22 15

(*) Call for Technology public presentation