mlsanityData to 7 Oct 2026

Morpho-MNIST: Quantitative Assessment and Diagnostics for Representation Learning

arXiv 1809.10780 ↗First posted 27 Sep 2018cs.LG stat.MLcited by 56

Not on the map. Why?

Abstract (arXiv)

Revealing latent structure in data is an active field of research, having introduced exciting technologies such as variational autoencoders and adversarial networks, and is essential to push machine learning towards unsupervised knowledge discovery. However, a major challenge is the lack of suitable benchmarks for an objective and quantitative evaluation of learned representations. To address this issue we introduce Morpho-MNIST, a framework that aims to answer: "to what extent has my model learned to represent specific factors of variation in the data?" We extend the popular MNIST dataset by adding a morphometric analysis enabling quantitative comparison of trained models, identification of the roles of latent variables, and characterisation of sample diversity. We further propose a set of quantifiable perturbations to assess the performance of unsupervised and supervised methods on challenging tasks such as outlier detection and domain adaptation. Data and code are available at https://github.com/dccastro/Morpho-MNIST.

Key facts. Cited by 56 papers in the map (data through 7 Oct 2026). First posted on arXiv 27 Sep 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 4; 2022: 8; 2023: 11; 2024: 13; 2025: 12; 2026: 8.

56 papers.

Sort:

By area:

  1. 2021FedBN: Federated Learning on Non-IID Features via Local Batch Normalizationcited by 565
  2. 2024Two Effects, One Trigger: On the Modality Gap, Object Bias, and Information Imbalance in Contrastive Vision-Language Modelscited by 55
  3. 2023High Fidelity Image Counterfactuals with Probabilistic Causal Modelscited by 53
  4. 2022D'ARTAGNAN: Counterfactual Video Generationcited by 27
  5. 2021Interventional Sum-Product Networks: Causal Inference with Tractable Probabilistic Modelscited by 22
  6. 2024Benchmarking Counterfactual Image Generationcited by 22
  7. 2023From Identifiable Causal Representations to Controllable Counterfactual Generation: A Survey on Causal Generative Modelingcited by 21
  8. 2024Causal Diffusion Autoencoders: Toward Counterfactual Generation via Diffusion Probabilistic Modelscited by 17
  9. 2023Feature Unlearning for Pre-trained GANs and VAEscited by 15
  10. 2022Deep Structural Causal Shape Modelscited by 11
  11. 2023Emerging Synergies in Causality and Deep Generative Models: A Surveycited by 10
  12. 2024Counterfactual Generative Modeling with Variational Causal Inferencecited by 10
38 more
  1. 2024Modular Learning of Deep Causal Generative Models for High-dimensional Causal Inferencecited by 9
  2. 2023Deep Backtracking Counterfactuals for Causally Compliant Explanationscited by 8
  3. 2024Learning Structural Causal Models through Deep Generative Models: Methods, Guarantees, and Challengescited by 8
  4. 2025Diffusion Counterfactual Generation with Semantic Abductioncited by 8
  5. 2021Turning Federated Learning Systems Into Covert Channelscited by 6
  6. 2024Semi-Supervised Learning for Deep Causal Generative Modelscited by 6
  7. 2022Explaining Image Classifiers Using Contrastive Counterfactuals in Generative Latent Spacescited by 4
  8. 2022GLANCE: Global to Local Architecture-Neutral Concept-based Explanationscited by 4
  9. 2023Towards objective and systematic evaluation of bias in artificial intelligence for medical imagingcited by 4
  10. 2024Causal Generative Explainers using Counterfactual Inference: A Case Study on the Morpho-MNIST Datasetcited by 4
  11. 2024Conditional Generative Models are Sufficient to Sample from Any Causal Effect Estimandcited by 4
  12. 2025Common Data Properties Limit Object-Attribute Binding in CLIPcited by 4
  13. 2025LeapFactual: Reliable Visual Counterfactual Explanation Using Conditional Flow Matchingcited by 4
  14. 2022Is More Data All You Need? A Causal Explorationcited by 3
  15. 2023A Causal Ordering Prior for Unsupervised Representation Learningcited by 3
  16. 2021Multiple Instance Learning with Auxiliary Task Weighting for Multiple Myeloma Classificationcited by 2
  17. 2022Counterfactual Generation Under Confoundingcited by 2
  18. 2024Benchmarking Dependence Measures to Prevent Shortcut Learning in Medical Imagingcited by 2
  19. 2025RBT4DNN: Requirements-based Testing of Neural Networkscited by 2
  20. 2026A Principled Approach to Unsupervised Anomaly Detectioncited by 2
  21. 2022Distributional Gaussian Processes Layers for Out-of-Distribution Detectioncited by 1
  22. 2023On Counterfactual Data Augmentation Under Confoundingcited by 1
  23. 2023Contrastive Attention Networks for Attribution of Early Modern Printcited by 1
  24. 2024Label Filling via Mixed Supervision for Medical Image Segmentation from Noisy Annotationscited by 1
  25. 2025Prompting Forgetting: Unlearning in GANs via Textual Guidancecited by 1
  26. 2025Intervening to Learn and Compose Causally Disentangled Representationscited by 1
  27. 2025GNN-based Unified Deep Learningcited by 1
  28. 2026Dataset Diversity Metrics and Impact on Classification Modelscited by 1
  29. 2022Oracle Guided Image Synthesis with Relative Queriescited by 0
  30. 2023Learning Confident Classifiers in the Presence of Label Noisecited by 0
  31. 2023Testing the Channels of Convolutional Neural Networkscited by 0
  32. 2024Natural Counterfactuals With Necessary Backtrackingcited by 0
  33. 2024Standardness Clouds Meaning: A Position Regarding the Informed Usage of Standard Datasetscited by 0
  34. 2025RefineSeg: Dual Coarse-to-Fine Learning for Medical Image Segmentationcited by 0
  35. 2025X-Node: Self-Explanation is All We Needcited by 0
  36. 2025Emerging Semantic Segmentation from Positive and Negative Coarse Label Learningcited by 0
  37. 2025FireGNN: Neuro-Symbolic Graph Neural Networks with Trainable Fuzzy Rules for Interpretable Medical Image Classificationcited by 0
  38. 2025Unsupervised Domain Adaptation via Content Alignment for Hippocampus Segmentationcited by 0

50 most cited of 56 All 56 (JSON)