HomeProgram

Deep Learning and Generative Models

Master the neural network architectures behind modern machine learning. This series of five short, stackable online courses teaches one focused skill at a time, from building your first PyTorch network, to training and evaluating deep models, to exploring the architectures and generative methods shaping the field today.

Program Information

5 Courses to complete program
Course Format: Self-Paced
Estimated: 10 weeks, 6-8 hours/week
Certificate: Program certificate on completion

Certificate Track

$300
Earn a verified certificate of completion
Access to this program & materials
Graded assignments & exams
MIT Open Learning certificate of completion

About this Program

Deep learning powers today’s most capable systems for recognition, prediction, and generation. Building these systems well depends on understanding the principles behind each architecture, making sound training choices, and using carefully designed experiments to measure meaningful progress.

  • This series of online courses from the MIT Electrical Engineering & Computer Science Department covers the foundations and practice of modern deep learning. Discover the science and craft of experimentation that will help you arrive at a model that actually works for its desired application.You will study neural networks, convolutional networks, transformers, representation learning, and generative models.
  • We introduce each topic intuitively, develop its mathematical foundations, and connect it to real data.
  • In hands-on PyTorch labs, you will build and compare models, choose meaningful baselines and evaluation metrics, test performance beyond the training conditions, and complete an original capstone experiment.
  • By the end of the series, you will understand how architecture and training choices shape model behavior, and be able to make sound design decisions from experimental evidence.

Show more

What you'll learn

  • Build and train neural networks in PyTorch, from linear layers and MLPs to deep residual networks
  • Diagnose training instead of guessing at it: initialization, normalization, optimizer choice, and what a loss curve that will not move is telling you
  • Match architecture to data with convolutional networks, transformers, and graph neural networks
  • Transfer a pretrained model to a new task, compress it for deployment, and evaluate it in and out of distribution
  • Learn representations without labels using autoencoders and contrastive methods, then measure what they encode
  • Build and evaluate generative models: VAEs, GANs, autoregressive models, and diffusion

Courses

To complete this program, you must take 5 required courses.

Required Courses

How you'll learn

  • Real-World Learning

    Learn from MIT faculty and experts who ground their teaching in real-world cases rather than mathematical models, making the material approachable for all.

  • Practical Application

    Apply your new knowledge with hands-on, practical exercises drawn from healthcare, sports, finance, sustainability, and more.

  • Learn On Demand

    Access all course content online with complete flexibility to study at your own pace.

  • AI-Enabled Support

    Deepen your understanding of the course material and get help on assignments from AskTIM, the AI assistant built by MIT researchers.

  • Stackable Credentials

    Earn an MIT Open Learning certificate at each milestone—module, course, and program—demonstrating your AI expertise. Available in paid courses only.

Prerequisites

This program is designed for:

  • Aspiring data scientists, machine learning engineers, software engineers moving into deep learning, and quantitative researchers.
  • Professionals who want to apply neural networks in technical or scientific work.
  • Advanced undergraduates, graduate students, and continuing education learners in computer science, data science, statistics, mathematics, engineering, and related quantitative fields.

Meet your instructors

Sara Beery

Assistant Professor, MIT Electrical Engineering and Computer Science

Dr. Sara Beery is the Homer A. Burnell Career Development Professor in the MIT Faculty of Artificial Intelligence and Decision-Making. She received her PhD in Computing and Mathematical Sciences at Caltech, where she was advised by Pietro Perona. Her research focuses on building computer vision methods that enable global-scale environmental and biodiversity monitoring across data modalities, tackling real-world challenges including geospatial and temporal domain shift, learning from imperfect data, fine-grained categories, and long-tailed distributions. Her work has been recognized with a Schmidt Sciences AI2050 Early Career Fellowship, an NSF CAREER Grant, the Amori Doctoral Prize, an Amazon AI for Science Fellowship, a PIMCO Data Science Fellowship, and an NSF GRFP. She partners with industry, nongovernmental organizations, and government agencies to deploy her methods in the wild worldwide. She works to increase access to AI skills through interdisciplinary capacity building and education, and was awarded the MIT EECS Outstanding Educator Award, has founded the AI for Conservation slack community, founded and directs the Workshop on Computer Vision Methods for Ecology, and co-leads the NSF/NSERC Global Center on AI and Biodiversity Change.