Skip to main content

Instructor guide

Workshop
In Progress

Module overview​

In this 4-week module, apprentices will learn comprehensive skills to develop, train, and optimize ML models that excel in real-world business environments. You'll explore various model architectures, fine-tune training processes, and implement advanced techniques to detect and address bias in complex datasets. These capabilities will help you develop robust ML models that deliver reliable results, solve various business challenges, and improve your organization's competitive advantage.

CompetencyLearning objectivesKSBsModel Development and Training Fundamentals- Analyze various ML model architectures and their suitability for different types of business problems, considering quality, efficiency, and validity principles.

  • Evaluate advanced training methodologies and their impact on model performance, security, and generalization across diverse datasets.
  • Develop frameworks for selecting and deploying appropriate ML models and hardware architectures based on computational requirements and available resources
  • K19: Evaluating Software Solutions
  • S12: Solving Business Problems with ML
  • B5: Operating in Technical Complexity Training Process Optimization- Analyze advanced performance metrics and their relevance to various business contexts, understanding how they reflect model effectiveness and business value.
  • Evaluate optimization techniques for ML model training, considering their impact on both model performance and computational efficiency.
  • Synthesize insights from performance metrics and testing results to make strategic decisions in ML model development and deployment, balancing business objectives with technical constraints.
  • Design and execute comprehensive model testing and tuning strategies that optimize performance for specific business objectives.
  • S7: Tuning and Testing Model Outputs
  • K10: Implementing ML for Data Engineering Identifying and Mitigating Sources of Bias- Analyze advanced techniques for model evaluation and validation, understanding their applicability across diverse ML problem domains and user groups.
  • Evaluate sources of algorithmic bias and their potential impacts on model fairness and ethics, considering dataset choices and methodological decisions.
  • Design and execute bias detection and mitigation strategies using explainable AI (XAI) techniques to enhance model transparency and fairness.
  • Develop protocols for continuous monitoring and adjustment of ML models to maintain fairness and performance across evolving user demographics and business needs.
  • K20: Identifying Errors and Algorithmic Bias
  • S20: Minimizing Bias in Models
  • K31: Best Practices in Software Development

Module breakdown​

TimingEvent (Links to Ariel Modules)PurposeWeek 1Kickoff WorkshopIntroduce the module, skills covered, and projectAsync unit(s):

Apply techniques to fine-tune and optimize the training process, enhancing model performance, computational efficiency, and model fit to deliver robust ML solutions.

Workshop 1: Hyperparameter TuningLearn how hyperparameter tuning drives real business impact by improving model performance and decision quality. Explore key techniques like grid search, support vector machines, AIC, and BIC—and how they help you fine-tune your models. Then, put it into practice by applying these methods to optimise a regression model.Week 2Async unit(s):

  • Identifying and Mitigating Sources of Bias Develop critical skills in bias detection, ensuring ML solutions are not only accurate but also fair and ethical across diverse user groups.Workshop 2: Handling Imbalanced DatasetsUnderstand the business value of tackling imbalanced classes to build fairer, more accurate models. Learn core techniques like oversampling, undersampling, SMOTE, and evaluation metrics like Kappa and AUC—and how they help manage bias. Finally, apply these methods to address class imbalance and improve model reliability in real-world scenarios.Module Wrap Up WorkshopIn this 45-minute workshop, you will recap what is covered in the Module, how they will apply those skills on the job, and what they will complete for their module project/milestone.Weeks 3-4Group Coaching Receive coach and peer support on the module projectProjectApply skills from the module to a real-world problem## How to prepare for live workshopsModule Kickoff Workshop####Workshop overview

In this 45-minute workshop, you will introduce apprentices to what is covered in the Module, how they can apply those skills on the job, and what they will complete for their module project/milestone.

By the end of this workshop, apprentices will be able to:

  • Identify the skills they will learn in the module and how they can apply them to their roles
  • Understand the elements of the module and how they fit together
  • Locate and review their module project/milestone

Delivery preparation​

  • Make a copy of the workshop slides (add share link, changing the word “edit” to “copy” in the URL).
  • Read the speaker notes in the workshop slides to understand what to cover during this workshop.
  • Ensure you understand what will be covered in this module and how the elements fit together.
  • Review the module project/milestone overview provided below.

Apprentice prerequisites​

There are no apprentice prerequisites for Kickoff workshops.

Workshop 1: Hyperparameter Tuning####Workshop overview

In this 1 hour workshop, you will review what was covered in async units 1 - 2. Then, apprentices will learn how hyperparameter tuning drives real business impact by improving model performance and decision quality. Explore key techniques like grid search, support vector machines, AIC, and BIC—and how they help you fine-tune your models. Then, put it into practice by applying these methods to optimise a regression model..

By the end of this workshop, apprentices will be able to:

  • Identify the business value of utilizing hyperparameter tuning techniques.
  • Define how the key concepts of grid search, support vector machine, AIC, and BIC enable hyperparameter tuning.
  • Apply hyperparameter tuning techniques to enhance a regression model.

Delivery preparation​

Apprentice prerequisites​

Apprentices should complete async units 1 -2 before attending this workshop. However, if apprentices did not complete the required async units, they are encouraged to attend.

Workshop 2: Handling Imbalanced Datasets####Workshop overview

In this 1 hour workshop, you will review what was covered in async units 3. Then, apprentices will understand the business value of tackling imbalanced classes to build fairer, more accurate models. Learn core techniques like oversampling, undersampling, SMOTE, and evaluation metrics like Kappa and AUC—and how they help manage bias. Finally, apply these methods to address class imbalance and improve model reliability in real-world scenarios..

By the end of this workshop, apprentices will be able to:

  • Identify the business value of using techniques for managing imbalanced class.
  • Define how the key concepts of imbalanced class, oversampling/undersampling, SMOTE, Kappa, and AUC support managing bias.
  • Apply techniques to address class imbalance and manage bias

Delivery preparation​

  • Make a copy of the workshop slides (add link).
  • Read the speaker notes in the workshop slides to understand what to cover during the workshop.
  • Review async units 3:Workshop 2: Handling Imbalanced Datasets
  • Add additional pre-delivery prep step(s):

Apprentice prerequisites​

Apprentices should complete async units 3 before attending this workshop. However, if apprentices did not complete the required async units, they are encouraged to attend.

Module Wrap-Up Workshop####Workshop overview

In this 45-minute workshop, you will review what was covered in the module and how those skills can drive impact. Additionally, you will preview what apprentices are expected to complete for their module project/milestone.

By the end of this workshop, apprentices will be able to:

  • Synthesize key takeaways of the module
  • Identify areas of impact that they can apply their module skills
  • Understand the requirements of their module project/milestone

Delivery preparation​

  • Make a copy of the workshop slides (add link).
  • Read the speaker notes in the workshop slides to understand what to cover during the workshop.
  • Review the Project overview provided below. Apprentice prerequisites

Apprentices should complete all async units and attend required workshops before the Wrap-Up module. However, if apprentices did not complete the required async units or attend previous workshops, they are encouraged to attend the Wrap-Up module.

Group Coaching####Session overview

In this 1 hour session, you will:

  • Review the project requirements
  • Describe how apprentices can ensure they pass the KSBs associated with the project
  • Pick 1 of the below discussion options:Lead a round of project presentations for all apprentices to share their project progress.
  • Lead apprentices in a reflection surrounding how to apply their skills within their roles and organizations.
  • Pick 1 of the below review/feedback options:Facilitate peer review of the project presentations and provide your own review of the projects.
  • Facilitate peer feedback on application opportunities and blockers.

Coach preparation​

  • Make a copy of the workshop slides (add link).
  • Read the speaker notes in the workshop slides to understand what to cover during the workshop.
  • Review the Project overview provided below.
  • Watch the how to create groups in the Ariel module video. Only create the group for your group coaching session when the session has started and you know who will be in attendance.

Apprentice prerequisites​

Apprentices should have made some progress on the project/milestone in order to actively discuss with their peers. Encourage apprentices to attend Group Coaching whether or not they have made progress on their project/milestone. If they have not made progress, they will hopefully be inspired by some of their peers.