Explainable and Responsible AI
About This Course
The field of Artificial Intelligence (AI) has been gaining increasing attention in the recent years. There has been widespread adoption of AI for use in almost all industries and for various purposes.
With the pervasive and widespread use of AI, there have been cases of AI hiccups or cases of AI bias reported in various applications. It has become inevitable that the spotlight is slowly shifting from how AI models can be applied to the need for Explainable & Responsible AI (XRAI).
The emphasis on XRAI is important in order to promote AI innovation and adoption, as well as to gain trust and confidence in the users of AI. Therefore, there is a need for further research, development and discussion in the AI community concerning the various aspects of XRAI. In the recent years, there is a rise in the creation of software tools and guidelines related to this area by the technology companies and regulators of various countries.
The type of AI covered in this course will include for example, discriminative AI, generative AI, machine learning, deep learning, etc.
At the organizational level, widespread use of Artificial Intelligence coding assistants introduces significant challenges across quality assurance, security, governance, and culture. Typical issues include inconsistent adoption and skills gap, quality and maintainability, and security vulnerabilities
To mitigate these risks, organisations require training that addresses both tool usage and team practices. Local enterprises are actively upskilling their engineering workforce. Specifically, teams need:
•Technical usage of Artificial Intelligence tools: hands-on configuration, prompting, and integrating AI code assistants with development environments, as well as guidance for reviewing, editing, and testing AI-generated code.
•Coding standards and documentation: best practices for code quality, internal documentation, and rationale for code choices to ensure maintainabilit
What You'll Learn
- Understand what is Explainable and Responsible AI and recognize AI adoption consideration factors
- Identify data factors that affect AI algorithm performance
- Assess the mitigating measures to prevent potential data issues
- Know how to achieve better AI project transparency
- Know how to assess the required level of human intervention
- Appraise and explain the requirements of AI model update and review
- Apply ethical considerations for designing or using AI solutions
- Know how to implement Responsible AI framework in different scenarios
- Know the working mechanism of a black box model and what are the tools for Explainable AI
- Assess the fairness of the AI model and explore bias reduction strategy
- Understand and analyze model vulnerabilities and know the mitigating factors for adversarial attacks
Entry Requirements
Good understanding of AI models and its usage
Have experience working with AI models or as user of AI models