ML@UVA homeJoin

Lectures

  • Wednesdays, 6:30–7:30 pm
  • Olsson Hall
  • No application needed

A lecture series on Wednesdays in Olsson Hall, starting from the basics and moving toward recent research. Slides are posted after each one.

Add the calendar

Fall 2026

11 lectures

Upcoming lectures

  • Recurrent Neural Networks

    John Kim

    RNNs, LSTMs, GRUs

    Slides

  • Transformers, Part 1: Attention

    Anmol Thapa

  • Transformers, Part 2: Training and LLMs

    Anmol Thapa

  • Generative Learning

    Shahzaib Tariq

    Autoencoders, variational autoencoders, generative adversarial networks, and diffusion.

  • Reinforcement Learning, Part 1

    Vrishak Vemuri

  • Reinforcement Learning, Part 2 (Tentative)

    Vrishak Vemuri

    Only if an additional session is needed.

Past lectures

  • Convolutional Neural Networks

    Joshua Yoo

    Slides

  • Neural Networks Continued and ML Science

    Shubhrangshu Debsarkar

  • Neural Networks

    Ishan Ajwani

  • Intro to ML, Part 2

    Decision trees, support vector machines, and perceptrons.

  • Intro to ML, Part 1

    Shubhrangshu Debsarkar

    Linear regression and logistic regression.

    Slides

Earlier semesters

Spring 202614 lectures

  • ML and Data Engineering in the Cloud

    Vrishak Vemuri and Ayaan Siddiqui

  • AI Agents

    John Kim and Joshua Yoo

    Topics: AI Agents, Tool Calling, Context and Memory, Retrieval-Augmented Generation, Model Context Protocol

    SlidesRecording

  • Linear Regression

    Utkarsh Goyal

    Topics: Linear Regression and Assumptions

    Slides

  • YOLO

    Anmol Thapa and Benjamin Hayworth

    Recording

  • Vision Transformer

    Logan Bradley and Ishan Ajwani

  • Alignment

    Seth Lifland

    Topics: Alignment, Reward Hacking, Deception, Mesa-optimization

    SlidesRecording

  • Mechanistic Interpretability

    Seth Lifland

    Topics: Decision Trees, Concept Bottleneck Models, Perturbation and Occlusion, Sparse Autoencoders

    SlidesRecording

  • Reinforcement Learning Part 2

    Seth Lifland

    Topics: Reinforcement learning on continuous state or action spaces

    SlidesRecording

  • Reinforcement Learning Part 1

    Seth Lifland

    Topics: Reinforcement learning on discrete state/action spaces, TD learning, Monte-Carlo RL

    SlidesRecording

  • Large Language Models

    Shubhrangshu Debsarkar and Seth Lifland

    Topics: LLM lifecycle, LLM limitations, Survey of Modern LLMs

    SlidesRecording

  • Transformers Part 3

    Seth Lifland

    Topics: Encoder vs Decoder

    Slides

  • Transformers Part 2

    Shubhrangshu Debsarkar and Seth Lifland

    Topics: Subspaces, Attention, Attention Pattern

    SlidesRecording

  • Transformers Part 1

    Shubhrangshu Debsarkar

    Topics: Intro to Transformer Architecture

    SlidesRecording

  • Review of Neural Networks

    Raffi Khondaker

    Topics: Neural Networks, Training, Forward and Backward Propagation, Over/Underfitting, Evaluation, Training Loops, Optimizers

    SlidesRecording

Fall 202510 lectures

  • Diffusion Models

    Raffi Khondaker

    Topics: Review of Image Generation, U-Net, Residual Encodings, Positional Encodings, Diffusion (DDPM), Latent Diffusion

    SlidesRecording

  • Generative Adversarial Networks

    Raffi Khondaker

    Topics: Generator / Discriminator, Generative Models, Probability

    SlidesRecording

  • Autoencoders & Variational Autoencoders (VAEs)

    Raffi Khondaker

    Topics: Encoder / Decoder, Variational Autoencoders, Denoising, Latent Spaces, Image Generation

    SlidesRecording

  • Using CS Server

    Raffi Khondaker

    Topics: GPUs, CUDA, HPC

    Slides

  • Optimization Techniques

    Raffi Khondaker

    Topics: Bias/Variance Tradeoff, Normalization, Regularization, Optimizers

    SlidesRecording

  • Recurrent Neural Networks (RNNs)

    Shubhrangshu Debsarkar

    Topics: RNNs, LSTMs, GRUs, Vanishing / Exploding Gradients | Prereqs: Previous lectures, Backpropagation

    SlidesRecording

  • Convolutional Neural Networks (CNNs)

    Raffi Khondaker

    Topics: CNNs | Prereqs: Neural Networks, Matrix Multiplication, Previous lectures

    SlidesRecording

  • Training Pipeline + Code

    Raffi Khondaker

    Topics: Training Loops, PyTorch, Optimizers | Prereqs: Previous lectures + Basic Python and Object Oriented Programming

    SlidesRecording

  • Training a Neural Network

    Raffi Khondaker

    Topics: Neural Networks, Training, Forward and Backward Propagation, Over/Underfitting, Evaluation

    SlidesRecording

  • Introduction to Deep Learning

    Raffi Khondaker

    Topics: Perceptrons, Neurons

    SlidesRecording

Spring 202512 lectures

  • Non-Deep ML

    Topics: Naive Bayes, Linear Regression, Logistic Regression, SVM, Decision Trees

    Slides

  • Graph Representation Learning

    Topics: Graphs, Node and Graph Embeddings, Traditional (non-deep) Approaches, Graph Neural Networks, GNN Architectures

    SlidesRecording

  • Optimization Techniques

    Topics: Bias/Variance Tradeoff, Normalization, Regularization

    SlidesRecording

  • Diffusion Models

    Topics: Diffusion, Latent Model Diffusion, Residual Connections

    SlidesRecording

  • Generative Adversarial Networks

    Topics: Generator/Discriminators, Probability, Mode Collapse, Minibatch Discrimination

    SlidesRecording

  • Mamba

    Topics: Mamba, structured space duality, causal linear attention, mamba 2

    SlidesRecording

  • State Space Models (SSMs)

    Topics: State Space Models, S4, S5, and a little on S6

    SlidesRecording

  • Large Language Models (Guest Lecture)

    Topics: LLM Architecture, Pretraining, Finetuning

    Slides

  • Reinforcement Learning Part 2

    Topics: Q-Learning, Deep Q Networks (DQN), Actor-Critic

    SlidesRecording

  • Reinforcement Learning Part 1

    Introduction to RL (non-deep) Topics: Markov Decision Process, Policy vs Value based, Monte-Carlo, TD Learning, Q Learning

    SlidesRecording

  • Rivanna + CS Server Usage

    An introduction to UVA computing, specifically how to use it for Machine Learning purposes Topics: tmux, SLURM, CUDA

    SlidesRecording

  • Neural Networks

    This is a quick refresher of deep learning + the former lectures

    SlidesRecording

Fall 20249 lectures

  • Transformers Part 3

    SlidesRecording

  • Transformers Part 2

    SlidesRecording

  • Transformers Part 1

    Topics: Vectors and Vector Spaces, Linear Transformations, Word Embeddings, Linear Representation Hypothesis, Transforming Embeddings

    SlidesRecording

  • Autoencoders & Variational Autoencoders (VAEs)

    Topics: Representation Learning, Probability and Normal Distributions, Encoders, Decoders, Autoencoders and Variational Autoencoders, Review of CNNs | Prereqs: CNNs

    Slides

  • Recurrent Neural Networks (RNNs)

    Topics: RNNs, LSTMs, GRUs, Vanishing / Exploding Gradients | Prereqs: Previous lectures, Backpropagation

    SlidesRecording

  • Convolutional Neural Networks (CNNs)

    Topics: CNNs | Prereqs: Neural Networks, Matrix Multiplication, Previous lectures

    SlidesRecording

  • Training Pipeline + Code

    Topics: PyTorch, Optimizers | Prereqs: Basic Python, Matrix Multiplication, Previous lectures

    SlidesRecording

  • Training a Neural Network

    Topics: Training, Forward and Backward Propagation, Over/Underfitting, Evaluation

    SlidesRecording

  • Introduction to Deep Learning

    Topics: Perceptrons, Neurons, Neural Networks

    SlidesRecording

Fall 202311 lectures

  • Autoencoders

    Raffi Khondaker

    Topics: Encoder / Decoder, Variational Autoencoders, Denoising, Latent Spaces, Image Generation

    Slides

  • Transformers Part 3

    Vainateya Rangaraju and Vyas Narasimhan

    Topics: Transformers and LLMs

    SlidesRecording

  • Transformers Part 2

    Vainateya Rangaraju and Vyas Narasimhan

    Topics: The transformer architecture

    SlidesRecording

  • Transformers Part 1

    Vainateya Rangaraju and Vyas Narasimhan

    Topics: Self-attention

    SlidesRecording

  • Recurrent Neural Networks (RNNs)

    Raffi Khondaker

    SlidesRecording

  • Convolutional Neural Networks (CNNs)

    Raffi Khondaker

    Slides

  • Review of Neural Networks

    Vainateya Rangaraju

    Slides

  • PyTorch

  • Code Demo

  • Overview of the ML Landscape

  • Introduction to Neural Networks

Reading Group30 papers

The reading group isn't meeting right now. These are the papers members presented and discussed.

Come to a lecture

No application or experience needed. Wednesdays, 6:30–7:30 pm, in Olsson Hall.

Join the Discord