Phase 3: Generative Vision Models · ~75 minutes · Python
Self-Supervised Vision — SimCLR, DINO, MAE
Labels are the bottleneck of supervised vision. Self-supervised pretraining removes them: learn visual features from 100M unlabelled images, fine-tune on 10k labelled ones.
What you will learn
- Trace the three major self-supervised families — contrastive (SimCLR), teacher-student (DINO), masked reconstruction (MAE) — and state what each one optimises
- Implement an InfoNCE loss from scratch and explain why a batch of 512 works but a batch of 32 fails
- Explain why MAE's 75% masking ratio is not arbitrary and how it differs from BERT's 15% for text
- Use DINOv2 or MAE ImageNet checkpoints for linear probing and zero-shot retrieval
Introduction
- Type: Learn + Build
- Languages: Python
- Prerequisites: Image Classification
- Time: ~75 minutes
Objective
Learning objectives
- Trace the three major self-supervised families — contrastive (SimCLR), teacher-student (DINO), masked reconstruction (MAE) — and state what each one optimises
- Implement an InfoNCE loss from scratch and explain why a batch of 512 works but a batch of 32 fails
- Explain why MAE's 75% masking ratio is not arbitrary and how it differs from BERT's 15% for text
- Use DINOv2 or MAE ImageNet checkpoints for linear probing and zero-shot retrieval
Unlock the full lesson
You've read the first 2 sections. The rest of this lesson covers The Problem, The Concept, Build, Check Yourself, Key Terms & Next — plus a hands-on lab, quiz, and project artifact.
Create a free account — the opening phases of 29 of 35 courses are free, no credit card.
Browse all courses · View pricing · DeVenture Academy