Second International Summer School on AI Foundations (ISSAIF'25)
at King's College London Wednesday 11th June - Friday 13th June 2025
Embedded Files
The summer school on Foundations of AI will allow participants to explore the fundamental methods and mathematical framework underpinning Artificial Intelligence (AI). It provides postgraduate students and computer science professionals with a deep understanding of AI techniques, including nature inspired methods, machine learning, neural networks, probability theory, and logic programming. Participants will gain insights to the potential of AI in revolutionising industries such as healthcare, finance, sustainability, and transportation. They will explore how AI can automate tasks, make predictions, and assist with decision-making.
The sessions will also emphasise the limitations and challenges of AI. Issues related to bias and fairness in AI algorithms, the need for responsible AI development, and the computational limitations of AI methods will also be covered. By the end of this summer school, participants will have a well-rounded understanding of AI, its immense potential, and the critical need to navigate its limitations responsibly. They will be equipped to contribute to the development and deployment of AI techniques in research and industry.
Prerequisites: undergraduate computer science degree or comparable - basic understanding of programming, logic, probability theory, algorithmics, representation of data. Participants are asked to bring a laptop to enjoy the full benefit of the course.
Schedule
Schedule
Day 1 - Wednesday 11th June
Model-driven AI: Logic Programming and SAT-solving
Day 1 - Wednesday 11th June
Model-driven AI: Logic Programming and SAT-solving
Welcome [9:30 AM - 10:00 AM]
Welcome [9:30 AM - 10:00 AM]
9:30 AM - 10:00 AM: Welcome and Overview
Brief introduction to the day's schedule.
Overview of AI and ML, their significance, and real-world applications.
Session 1: Logic Programming (Part 1) [10:00 AM - 10:45PM]
Session 1: Logic Programming (Part 1) [10:00 AM - 10:45PM]
10:00 AM - 10:45 AM: Introduction to Logic Programming
Introduction/recap of propositional and predicate logic.
Horn clauses and clausal resolution
Break: 10:45 AM - 11:00 AM
Session 2: Logic Programming (Part 2) [11:00 AM - 11:45 PM]
Session 2: Logic Programming (Part 2) [11:00 AM - 11:45 PM]
11:00 AM - 11:45 PM: Logic Programming in Prolog
Introduction to Prolog for Logic Programming
Lunch Break [11:45 PM - 1:00 PM]
Lunch Break [11:45 PM - 1:00 PM]
Session 3: SAT-solving (Part 1) [1:00 PM - 1:45 PM]
Session 3: SAT-solving (Part 1) [1:00 PM - 1:45 PM]
1:00 PM - 1:45 PM: Introduction to SAT-solving
The complexity of SAT
Introduction to the DPLL algorithm for SAT-solving
Break: 1:45 PM - 2:00 PM
Session 4: SAT-solving (Part 2) [2:00 PM - 2:45 PM]
Session 4: SAT-solving (Part 2) [2:00 PM - 2:45 PM]
2:00 PM - 2:45 PM: Translating problems into SAT
Introduction to the PySAT library
Procedurally generating and solving SAT problems
Break: 2:45 PM - 3:00 PM
Session 5: Finite Automata [3:00 PM - 3:45 PM]
Session 5: Finite Automata [3:00 PM - 3:45 PM]
3:00 PM - 3:45 PM: Introduction to Finite Automata
Break: 3:45 PM - 4:00 PM
Session 6: Lab [4:00 PM - 5:00PM]
Session 6: Lab [4:00 PM - 5:00PM]
Coding exercises
Day 2 - Thursday 12th June
Nature-Inspired AI: Randomness and Meta-Heuristics
Day 2 - Thursday 12th June
Nature-Inspired AI: Randomness and Meta-Heuristics
Session 1: Introduction to Randomized Algorithms [10:00 AM - 10:45 AM]
Session 1: Introduction to Randomized Algorithms [10:00 AM - 10:45 AM]
Definition and importance of randomized algorithms.
Randomized algorithms vs. deterministic algorithms.
Examples of how to design and analyse randomized methods.
Break: 10:45 AM - 11:00 AM
Session 2: Nature inspired Meta-Heuristics [11:00 AM - 11:45 AM]
Session 2: Nature inspired Meta-Heuristics [11:00 AM - 11:45 AM]
Introduction to Local Search for combinatorial optimisation.
Sequential vs distributed methods.
Examples of the design of search methods and performance criteria.
Lunch Break: 11:45 AM - 1:00 PM
Lunch Break: 11:45 AM - 1:00 PM
Session 3: Modelling Stochastic Processes with Markov Chains [1:00 PM - 1:45 PM]
Session 3: Modelling Stochastic Processes with Markov Chains [1:00 PM - 1:45 PM]
Definition and basic properties of Markov chains.
Transition probabilities and transition matrices.
Analysing Simulated Annealing with varying cooling schedules.
Break: 1:45 PM - 2:00 PM
Session 4: Real-world Applications [2:00 PM - 2:45 PM]
Session 4: Real-world Applications [2:00 PM - 2:45 PM]
Markov chain Monte Carlo (MCMC) methods.
Applications in machine learning, finance, and biology.
Practical examples and implementations.
Break: 2:45 PM - 3:00 PM
Session 5: Guest Speaker [3:00 PM - 3:45 PM]
Session 5: Guest Speaker [3:00 PM - 3:45 PM]
TBC
Break: 3:45 PM - 4:00 PM
Lab Session: [4:00 PM - 5:00 PM]
Lab Session: [4:00 PM - 5:00 PM]
Exercises and programming examples.
Model evaluation.
Data processing.
Day 3 - Friday 13th June 2024
Day 3 - Friday 13th June 2024
Data-driven AI: Clustering and Neural Networks
Data-driven AI: Clustering and Neural Networks
Morning Session: Foundations of ML [10:00 AM - 10:45 PM]
Morning Session: Foundations of ML [10:00 AM - 10:45 PM]
Introduction to Supervised Learning: Labelled data, regression, classification, examples.
Introduction to Unsupervised Learning: Data without labels, finding patterns, examples.
Introduction to Reinforcement Learning: Learning through interaction with an environment, reward systems, examples.
Modelling uncertainty.
Break: 10:45 AM - 11:00 AM
Session 2: Clustering - Unsupervised Learning [11:00 AM - 11:45 PM]
Session 2: Clustering - Unsupervised Learning [11:00 AM - 11:45 PM]
Basics of Clustering: Definition, importance, and applications in various fields.
Popular Clustering Algorithms: Overview of K-means, hierarchical clustering, and DBSCAN.
Hands-On Exercise: Simple clustering exercise using a tool like Python with libraries such as scikit-learn. Participants will apply a clustering algorithm to a dataset and interpret the results.
11:45 PM - 1:00 PM: Lunch Break
11:45 PM - 1:00 PM: Lunch Break
Session 3: Neural Networks - Supervised Learning 1 [1:00 PM - 1:45 PM]
Session 3: Neural Networks - Supervised Learning 1 [1:00 PM - 1:45 PM]
Basics of Neural Networks: Neurons, weights, biases, and activation functions.
Architecture of Neural Networks: Layers, input, hidden layers, output, how they mimic the brain.
How Neural Networks Learn: Forward propagation, loss functions, backpropagation, and gradient descent.
Break: 1:45 PM - 2:00 PM
Session 4: Neural Networks - Supervised Learning 2 [2:00 PM - 2:45 PM]
Session 4: Neural Networks - Supervised Learning 2 [2:00 PM - 2:45 PM]
Universal approximation theorem
Importance of non-linearity
Break: 2:45 PM - 3:00 PM
Session 5: Reinforcement Learning [3:00 PM - 3:45 PM]
Session 5: Reinforcement Learning [3:00 PM - 3:45 PM]
Markov Decisions Processes (MDPs)
Multi-Armed Bandits (Exploration vs Exploitation)
Deep Q-learning
Break: 3:45 PM - 4:00 PM
Session 5: Lab [4:00 PM - 4:45 PM]
Session 5: Lab [4:00 PM - 4:45 PM]
Coding exercises
Wrap-Up and Takeaways [4:45 PM - 5:00 PM]
Wrap-Up and Takeaways [4:45 PM - 5:00 PM]
Recap of the day's key points.
Open floor for final questions and discussions.
Guidance on further resources for deepening knowledge in ML and AI.
This schedule is designed to offer a balance between theoretical understanding and practical application, providing a solid foundation in key areas of AI and ML. Adjustments may be necessary based on the audience's background and interests, as well as logistical considerations.
Organisers
Organisers
Dr Kathleen Steinhöfel
Dr Kathleen Steinhöfel
Kathleen is a Reader in Computer Science at the Department of Informatics, King’s College London, and member of the Algorithms and Data Analysis group. She completed her Ph.D. degree (Dr rer nat) at the Technical University Berlin with suma cum laude. Before joining King’s College London, she conducted postdoctoral research at ETH Zürich, CUHK, and MIT. Her research interests are in stochastic algorithms, combinatorial optimisation, algorithmic learning, neural networks and algorithmic bioinformatics.
Dr Christopher Hampson
Dr Christopher Hampson
Chris is a Senior Lecturer in Computer Science Education in the Department of Informatics at King's. His background is in Mathematics, graduating from The University of Manchester before joining King's in 2011 to commence his PhD research in Computer Science. His research interests lie in logic and formal reasoning, algorithm design, and computational complexity.
Dr Frederik Mallmann-Trenn
Dr Frederik Mallmann-Trenn
Frederik is a Senior Lecturer in Data Science at King's College London and head of the Algorithms and Data Analysis group. He did his postdoc at MIT and directs the Random Lab. He has a worked on the theory behind AI and supervised close to 50 theses, and his engagement with industry professionals ensures a curriculum that's both academically solid and industry-relevant.
Venue
Venue
King's College London, Strand Campus
King's College London, Strand Campus
Room: BH.(S).4.04
Room: BH.(S).4.04
The event will be hosted at King's College London King's College London's Strand Campus is wonderfully located in the center of London, making it a vibrant spot. It's right in the heart of the city's theater district, so you're never far from catching a fantastic show. The bustling atmosphere of the area, combined with the scenic beauty of the nearby Thames, really makes it a lively place to study and hang out. It’s not just about books and lectures here—there's a whole city to explore right on your doorstep!
The event will be hosted at King's College London King's College London's Strand Campus is wonderfully located in the center of London, making it a vibrant spot. It's right in the heart of the city's theater district, so you're never far from catching a fantastic show. The bustling atmosphere of the area, combined with the scenic beauty of the nearby Thames, really makes it a lively place to study and hang out. It’s not just about books and lectures here—there's a whole city to explore right on your doorstep!
Registration and Fees:
Registration and Fees:
Please register here.
The registration includes the live in-person tutorials and the material provided. Accommodation and food are not included.
Cost for all three days:
£450 Early Bird (until 28.04.25)
£540 Regular registration (from 01.06.25)
Payment via Eventbrite opens 1.03.25. Refund policy until 48h before the event: Full refund minus 20 pound admin fee.
Contact: frederik.mallmann-trenn@kcl.ac.uk
Page updated
Google Sites
Report abuse