Internal seminars

Regular knowledge-sharing sessions among lab members on the latest research, methods and industry trends.

  1. 01
    LLMs

    DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models

    • RLHF fundamentals: review of Proximal Policy Optimization (PPO) and its traditional role in alignment
    • Enter GRPO (Group Relative Policy Optimization): mechanics and how it removes the separate critic model
    • PPO vs GRPO: memory efficiency, training stability & computational overhead
  2. 02
    LLMs

    Understanding Self-Attention Variants in Modern LLMs

    • Recap: Query, Key and Value in self-attention
    • Causal vs bidirectional self-attention
    • Single-head vs multi-head attention
    • Multi-Head Attention (MHA)
    • Multi-Query Attention (MQA)
    • Grouped-Query Attention (GQA)
    • Cross-attention vs self-attention
  3. 03
    LLMs

    Large Language Models for Combinatorial Optimization

    • Overview of the review paper "Large Language Models for Combinatorial Optimization: A Systematic Review"
    • Introduction to optimization problems
    • Introduction to combinatorial optimization problems
    • Large language models for combinatorial optimization
  4. 04
    LLMs

    Understanding LLM Evaluation: Hands on Implementation

    • Recap of the 4 paradigms of LLM evaluation & motivation
    • GSM8K: verifier demo
    • HumanEval / Pass@k: execution demo
    • MT-Bench: LLM-as-a-Judge
  5. 05
    LLMs

    Agentic AI in the AI Era

    • Concept & key characteristics of agentic AI
    • Evolution from traditional AI and generative AI to agentic AI
    • Main components: planning, memory, reasoning & tool use
    • Benefits, limitations, risks & ethical concerns of autonomous agents
    • Future trends in human-AI collaboration and multi-agent systems
  6. 06
    LLMs

    Fine Tuning LLMs

    • Introduction to LLM fine-tuning
    • Fine-tuning methods (Full, LoRA, QLoRA, PEFT)
    • LLM training pipeline
    • Hardware & optimization
    • Evaluation & benchmarking
    • Deployment & inference
  7. 07
    LLMs

    Understanding Large Language Models (LLMs)

    • What LLMs are & common use cases
    • Tokens & next-token prediction
    • Pretraining, datasets & scaling laws
    • Context window & compute requirements
  8. 08
    Quantum

    Basic of Quantum Computing (Part 2)

    • Quantum measurement mechanics
    • Device ecosystems & QPUs
    • Quantum software stack (Qiskit)
    • Hardware noise & decoherence
  9. 09
    LLMs

    Understanding the 4 Main Approaches to LLM Evaluation

    • Foundations of LLM evaluation
    • Multi-dimensional quality & failure modes
    • Benchmark & verifier-based evaluation
    • Human preference & LLM-as-a-judge
    • Paradigm trade-offs
  10. 10
    ML Fund.

    Hyperparameter Tuning: Controlling and Improving Neural Network Training

    • What are hyperparameters?
    • Learning rate, batch size & epochs
    • Tuning strategies
  11. 11
    ML Fund.

    Regularization Techniques: Preventing Overfitting and Improving Generalization

    • Overfitting & underfitting
    • Dropout, L1 & L2 regularization
    • Generalization strategies
  12. 12
    ML Fund.

    How Neural Networks Improve: Optimization Algorithms and Learning Dynamics

    • Gradient descent variants
    • Adam, SGD & momentum
    • Learning dynamics & convergence
  13. 13
    ML Fund.

    Backpropagation: How Neural Networks Learn Through Gradient Flow

    • Backpropagation algorithm
    • Gradient flow & chain rule
    • Weight updates
  14. 14
    Quantum

    Basic of Quantum Computing (Part I)

    • Introduction to quantum computing
    • Mathematical language of qubits
    • Superposition, entanglement & interference
    • Quantum circuits & operations
  15. 15
    AI Tools

    N8N Workflow Automation: From Fundamentals to AI Integration

    • N8N architecture, nodes & workflows
    • Hands-on pipelines
    • LLM & RAG integration
    • Real-world use cases
  16. 16
    ML Fund.

    Activation Functions and Loss Functions: Enabling Learning in Neural Networks

    • Activation functions
    • Loss functions
    • Neural network learning
  17. 17
    ML Fund.

    Model Architecture and Forward Propagation: How Neural Networks Are Structured and Compute Predictions

    • Neural network architecture
    • Forward propagation
    • Prediction computation
  18. 18
    AI Tools

    Anthropic Products and Features

    • Anthropic background
    • Claude model use cases
    • Claude web interface
    • Claude Code CLI & desktop
    • Key Claude Code features
  19. 19
    AI Tools

    Understanding Gemini AI Ecosystem and Use Cases

    • Gemini multimodal model & context window
    • Deep Research & Gemini Live
    • Image/video generation (Imagen, Veo)
    • Developer tools (Firebase Studio, AI Studio, Gemini CLI)
    • NotebookLM & Google Vids
  20. 20
    AI Tools

    OpenAI Products and Features

    • OpenAI introduction
    • ChatGPT products & use cases
    • Sora & Codex
    • OpenAI platform & developer ecosystem
  21. 21
    ML Fund.

    Foundations of Building Machine Learning Models

    • Problem definition
    • Data handling & preprocessing
  22. 22
    LLMs

    PAPER : Attention is all you need

    • Transformer overview & architecture
    • Encoder
    • Methodology & results
    • Model behavior analysis
  23. 23
    Vision

    Deep Dive into Object Detection : How models see "what"and "where"

    • Object detection overview & model output
    • Two-stage & one-stage detectors
    • Backbone / neck / head architecture
    • Evaluation metrics
  24. 24
    Research

    Mastering Data Storytelling and Visualization in Power BI

    • Plot types & interpretation
    • Power BI
    • Data storytelling for publications
  25. 25
    LLMs

    Understanding fundamentals of Prompt Engineering

    • Prompt engineering definition
    • Mechanics of LLM prompting
    • Core components of a prompt
    • Chain-of-Thought (CoT)
  26. 26
    LLMs

    An Introduction to NL2SQL

    • NL2SQL core ideas & pipeline
    • Models & schema encoding
    • Applications & challenges
  27. 27
    Vision

    Understanding Computer Vision and Applications

    • Computer vision overview
    • Image classification
    • Object detection & tracking
    • Object segmentation
    • Pose estimation
  28. 28
    Research

    Understanding Journal Indexing and Publication Metrics

    • Journal indexing databases
    • Impact Factor vs. CiteScore
    • Author-level metrics (h-index)
    • Google Scholar
  29. 29
    Research

    Understanding Basic LaTeX

    • What is LaTeX?
    • Essential commands
    • Structuring a document
    • Citations & bibliography
    • Creating a PDF
  30. 30
    Research

    Understanding Submission and Revision Processes

    • Choosing a journal
    • Formatting for submission
    • Peer review process
    • Responding to reviewer comments
    • Final proofreading & publication
  31. 31
    Research

    Citation and Reference Management

    • Citation styles
    • Manual citation
    • Reference management systems
    • Creating a bibliography
  32. 32
    Research

    Understanding Research Paper Structure

    • Research paper structure overview
    • Description of each section
    • Sample journal walkthrough