Explainable & Trustworthy Artificial Intelligence
20 March 2023 – 19 June 2023
Artificial Intelligence (AI) has come a long way since its first use and application many decades ago. The use of AI and Machine Learning have seen an immense uptake in the 21st century. The techniques developed in the domain were and are successfully applied to a wide variety of problems, both in academia, private and public industry. As this domain became more and more established in recent years, new challenges arose.
Artificial Intelligence nowadays are complex and sophisticated algorithms that sometimes make it difficult for the humans to understand and interpret the decisions or suggestions of
the AI system.
Explainable AI puts the following properties on the foreground to deliver trust:
- Gaining trust by explaining for example the characteristics of AI output.
- By explaining an AI technique understanding will increase, allowing to investigate if the technique can be transferred to another domain or problem.
- Informing a user about the workings of an AI model so that there is no misinterpretation.
- Confidence of users can be established by using AI models that are explainable, stable but also robust.
- When explaining AI models issues concerning privacy awareness come into play. Private data should not be exposed by the models.
- It is important that actions can be explained. How have we come to specific outcomes and how could we change them?
- Nowadays, a wide variety of people from different background come into contact AI, it is important that they all understand why the system is behaving in such a manner and offer explanations tailored to their needs.
This course is a collaboration between UGain and VAIA.
The lessons are intended for anyone who has a good professional familiarity with computer science and who would like to get more insights in techniques that can be applied to achieve explainable and trustworthy Artificial Intelligence. Participants have completed a higher education in computer science or have acquired an equivalent experience.
Participants have programming experience with Python or a related programming language.
The lessons can be followed onsite in the UGain classroom (limited to 35 participants) or online live & on demand (unlimited).
To receive a certificate, one should attend all the lessons and succeed for the final exam. The exam will take place on 25 September 2023.
Scientific Coordinationdr. Femke De Backere, Department of Information Technology, Ghent University – imec & VAIA
- prof. Tijl De Bie, Department of Electronics and Information Systems, Ghent University
- dr. Arne Gevaert, Department of Applied Mathematics, Computer Science & Statistics, Ghent University
- ir. Mathias Leys, ML6
- prof. Femke Ongenae, Department of Information Technology, Ghent University
- dr. Daniel Peralta Cámara, Department of Information Technology, Ghent University
- dr. Jonathan Peck, Department of Applied Mathematics, Computer Science & Statistics, Ghent University
- dr. Jan Van Looy, ML6
- prof. Sofie Van Hoecke, Department of Electronics and Information Systems, Ghent University
- dr. Roel Wuyts, imec
In this first lesson, we give a short recap of the basics, followed by the explanation of some general terms that are used in the domain of explainable and trustworthy Artificial Intelligence. This introduction will end with the definition of the challenges within this domain.
- Recap the basics: AI, ML and statistics
- Different types of ML: white-box & black-box
- Interpretability vs Explainability
- Human Uncertainty vs Model Uncertainty
Teachers: Femke Ongenae & Sofie Van Hoecke
Date: 20 March 2023
Machine learning systems help us connect cause and effect in complex data sets. How we can make those interpretations depends on the type of algorithm used. The level of interpretability we can reach will also differ depending on whether we look at how the model predicts in general, versus how a specific prediction of the model was computed. When interpretability is good enough, we can use hypothetical questions like “What if instead this would be the case?”, adding to the usefulness of your system. This lesson explains how to increase and measure those interpretabilities.
- Global Interpretability
- Local Interpretability
- Model distillation
- Dependency plots
- Evaluation methods
Date: 27 March 2023
While black box models offer higher accuracy, white box models are easier to explain and to interpret, unfortunately this leads to a lesser predictive capacity. In the area of white box models, several different approaches will be highlighted:
- Linear Regression
- Decision Trees and Rule Sets
- Generalized Additive Models (GAMs)
Teacher: Daniel Peralta Cámara
Date: 17 April 2023
To help explain how a neural network reached a certain conclusion, certain visualizations of its reasoning can be useful. One type of visualization is a heatmap showing which areas of a photo contribute most to how a system has labeled it. These visualizations are explained in this session.
- Convolutional Neural Networks (CNNs) & interpretation
- Other gradient-based methods
Teacher: Arne Gevaert
Date: 24 April 2023
The oldest forms of machine learning entail rule engines that were hand programmed. Newer forms entail algorithms searching for connections themselves. The first are great in explaining how they reach their conclusions. The latter sometimes give superior predictions, being a lot less brittle, but lack that explainability. To get the best of both worlds, these approaches are sometimes combined. Moreover, allowing an expert to guide a machine learning system can sometimes lead to yet again superior predictions.
- Data-driven vs expert-based approaches
- Finding synergies in data-driven and expert-based approaches
- Combining expert knowledge and machine learning
Teachers: Femke Ongenae & Sofie Van Hoecke
Date: 8 May 2023
The output of a machine learning system depends on the data used as input. Often the needed amount and structure of that data is overlooked. However, machine learning systems can be combined to generate additional data or to finetune each other. Nevertheless, malicious additions to your training data can corrupt your system and even a well-trained system can be deceived. This lesson explains these issues and what you can do about them.
- Data quality & quantity
- Adversarial learning
- Poisoning attacks & defenses
- Learning theory
Teacher: Jonathan Peck
Date: 15 May 2023
Training machine learning systems can be done before use, i.e. when training it on a stack of pictures first and asking it to make sense of new pictures later. However, it can also be done during use. In the latter scenario the system gets updated whilst it is being used. Sometimes this is necessary because training data is (partially) becoming available after commissioning of the system. Sometimes a system is pretrained on one dataset and the developer wants to retrain the system in order to solve another but related problem, i.e. using a machine vision system that is trained to detect cats to now detect dogs. The developer thus leverages the effort put into the training of the earlier system, hence requiring less training time for the novel system. These and other relations between datasets, their application in training models and the problems we solve with those will be explained in this lesson.
- Online learning
- Change detection
- Transfer learning & domain adaptation (foundation models)
Teachers: Mathias Leys & Jan Van Looy
Date: 22 May 2023
When training machine learning systems, the training data can be biased, leading to unwanted outcomes, i.e. an HR system trained on old hospital personnel data now stating that women might be unlikely good candidates for doctor positions. This session will explain these issues, how to avoid them, how to measure bias and what the limitations of avoiding it are.
- Various notions of fairness & impossibility theorem
- Different types of bias & methods to debias
- Ethical guidelines
- Learning fair models
- Uncovering model bias
Teacher: Tijl De Bie
Date: 5 June 2023
Sometimes the quality of machine learning system outputs and privacy are at odds and need to be balanced. However, there are techniques that allow the training of machine learning systems on privacy sensitive data, without exposing the data itself. Those techniques and relevant regulation on these practices are explained in this session.
- Differential privacy
Teacher: Tijl De Bie
Date: 12 June 2023
During the last session, some specific use cases in the domain of Explainable and Trustworthy AI will be discussed.
Teachers: Tijl De Bie, Mathias Leys, Femke Ongenae, Sofie Van Hoecke & Roel Wuyts
Date: 19 June 2023
OnsiteThe fee for onsite participation (in the UGain classroom) is 1.650 euro.
This includes the tuition fee, course notes, access to the digital e-learning environment, soft drinks, coffee and sandwiches.
OnlineThe fee for online participation only is 1.400 euro.
This includes tuition fee and online access to the live sessions and the digital e-learning environment with digital course notes and recorded lessons.
All invoices are due in thirty days. All fees are exempt from VAT.
ReductionWhen a participant of a company subscribes for the complete course, a reduction of 20% is given to all additional subscriptions from the same company. In that case, only one invoice is issued per company.
Cancellation policyCancellation must be done in writing. Our cancellation conditions can be consulted on www.ugain.ugent.be/cancellation
Training vouchersGhent University accepts payments by KMO-portefeuille (www.kmo-portefeuille.be; authorisation ID: DV.O103194).
Vlaams Opleidingsverlof (VOV)This course has too few contact hours to qualify for VOV.