Author: Yusuf Brima

Seminar Outline

Week 1: Introduction to Representation Learning

• Overview of the similarities and differences between machine-based and biological representation learning
• Basic concepts of machine-based and biological representation learning

Week 2: Autoencoders

• The architecture and working of autoencoders
• Similarities between autoencoders and biological neural networks, such as the role of feedback loops in both systems

Week 3: Deep Generative Models

• The concept of deep generative models
• Comparisons between deep generative models and biological neural networks, such as the importance of hierarchical representations

Week 4: Deep Reinforcement Learning

• The concept of deep reinforcement learning
• Connections between reinforcement learning and biological learning, such as the role of dopamine in reward-based learning

Week 5: Metric Learning

• The concept of metric learning
• Comparisons between metric learning in machine learning and metric learning in biological systems, such as the role of distance metrics in neural coding

Week 6: Siamese Networks

• The architecture and working of Siamese networks
• Similarities between Siamese networks and biological systems, such as the role of neural synchronization in similarity encoding

Week 7: Graph Neural Networks

• The architecture and working of graph neural networks
• Connections between graph neural networks and biological neural networks, such as the role of neural networks in encoding and decoding information in graph structures

Week 8: Contrastive Learning

• The concept of contrastive learning
• Connections between contrastive learning and biological learning, such as the role of similarity-based learning in neural coding

Week 9: Self-Supervised Learning

• The concept of self-supervised learning
• Comparisons between self-supervised learning in machine learning and self-supervised learning in biological systems, such as the role of predictive coding in neural learning

Week 10: Federated Learning

• The concept of federated learning
• Connections between federated learning and biological learning, such as the role of distributed learning and communication in neural systems

Week 11: Meta-Learning

• The concept of meta-learning
• Comparisons between meta-learning in machine learning and meta-learning in biological systems, such as the role of adaptation and plasticity in neural networks

Week 12: Multi-Task Learning

• The concept of multi-task learning
• Connections between multi-task learning and biological learning, such as the role of flexible and adaptive neural networks in performing multiple tasks

Week 13: Active Learning

• The concept of active learning
• Connections between active learning and biological learning, such as the role of attention and curiosity in neural learning

Week 14: Recap and Discussion

• Recap of the key concepts and connections between machine-based and biological representation learning
• Discussion of the future directions and potential applications of representation learning, including the potential for bridging the gap between machine-based and biological representation learning