Trak: Attributing model behavior at scale
The goal of data attribution is to trace model predictions back to training data. Despite a long
line of work towards this goal, existing approaches to data attribution tend to force users to …
line of work towards this goal, existing approaches to data attribution tend to force users to …
Describing differences in image sets with natural language
How do two sets of images differ? Discerning set-level differences is crucial for
understanding model behaviors and analyzing datasets yet manually sifting through …
understanding model behaviors and analyzing datasets yet manually sifting through …
Facts: First amplify correlations and then slice to discover bias
S Yenamandra, P Ramesh… - Proceedings of the …, 2023 - openaccess.thecvf.com
Computer vision datasets frequently contain spurious correlations between task-relevant
labels and (easy to learn) latent task-irrelevant attributes (eg context). Models trained on …
labels and (easy to learn) latent task-irrelevant attributes (eg context). Models trained on …
Similarity of neural network models: A survey of functional and representational measures
Measuring similarity of neural networks to understand and improve their behavior has
become an issue of great importance and research interest. In this survey, we provide a …
become an issue of great importance and research interest. In this survey, we provide a …
Understanding the detrimental class-level effects of data augmentation
Data augmentation (DA) encodes invariance and provides implicit regularization critical to a
model's performance in image classification tasks. However, while DA improves average …
model's performance in image classification tasks. However, while DA improves average …
Bias-to-text: Debiasing unknown visual biases through language interpretation
Biases in models pose a critical issue when deploying machine learning systems, but
diagnosing them in an explainable manner can be challenging. To address this, we …
diagnosing them in an explainable manner can be challenging. To address this, we …
DiG-IN: Diffusion Guidance for Investigating Networks-Uncovering Classifier Differences Neuron Visualisations and Visual Counterfactual Explanations
While deep learning has led to huge progress in complex image classification tasks like
ImageNet unexpected failure modes eg via spurious features call into question how reliably …
ImageNet unexpected failure modes eg via spurious features call into question how reliably …
Decomposing and editing predictions by modeling model computation
How does the internal computation of a machine learning model transform inputs into
predictions? In this paper, we introduce a task called component modeling that aims to …
predictions? In this paper, we introduce a task called component modeling that aims to …
The journey, not the destination: How data guides diffusion models
Diffusion models trained on large datasets can synthesize photo-realistic images of
remarkable quality and diversity. However, attributing these images back to the training data …
remarkable quality and diversity. However, attributing these images back to the training data …
Data Debiasing with Datamodels (D3M): Improving Subgroup Robustness via Data Selection
Machine learning models can fail on subgroups that are underrepresented during training.
While techniques such as dataset balancing can improve performance on underperforming …
While techniques such as dataset balancing can improve performance on underperforming …