Contrastive Adapters for Foundation Model Group Robustness
- URL: http://arxiv.org/abs/2207.07180v1
- Date: Thu, 14 Jul 2022 19:40:55 GMT
- Title: Contrastive Adapters for Foundation Model Group Robustness
- Abstract summary: Existing methods to improve robustness require retraining, which can be prohibitively expensive on large foundation models.
We propose contrastive adapting, which trains adapters with contrastive learning to bring sample embeddings close to both their ground-truth class embeddings and other sample embeddings in the same class.
Our approach consistently improves group robustness, raising worst-group accuracy by 8.5 to 56.0 pp over zero-shot.
- Score: 11.961187296775185
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: While large pretrained foundation models (FMs) have shown remarkable
zero-shot classification robustness to dataset-level distribution shifts, their
robustness to subpopulation or group shifts is relatively underexplored. We
study this problem, and find that FMs such as CLIP may not be robust to various
group shifts. Across 9 robustness benchmarks, zero-shot classification with
their embeddings results in gaps of up to 80.7 percentage points (pp) between
average and worst-group accuracy. Unfortunately, existing methods to improve
robustness require retraining, which can be prohibitively expensive on large
foundation models. We also find that efficient ways to improve model inference
(e.g., via adapters, lightweight networks with FM embeddings as inputs) do not
consistently improve and can sometimes hurt group robustness compared to
zero-shot (e.g., increasing the accuracy gap by 50.1 pp on CelebA). We thus
develop an adapter training strategy to effectively and efficiently improve FM
group robustness. Our motivating observation is that while poor robustness
results from groups in the same class being embedded far apart in the
foundation model "embedding space," standard adapter training may not bring
these points closer together. We thus propose contrastive adapting, which
trains adapters with contrastive learning to bring sample embeddings close to
both their ground-truth class embeddings and other sample embeddings in the
same class. Across the 9 benchmarks, our approach consistently improves group
robustness, raising worst-group accuracy by 8.5 to 56.0 pp over zero-shot. Our
approach is also efficient, doing so without any FM finetuning and only a fixed
set of frozen FM embeddings. On benchmarks such as Waterbirds and CelebA, this
leads to worst-group accuracy comparable to state-of-the-art methods that
retrain entire models, while only training $\leq$1% of the model parameters.
Related papers
- Too Correct to Learn: Reinforcement Learning on Saturated Reasoning Data [55.84428098924793]
We propose Constrained Uniform Top-K Sampling (CUTS), a parameter-free decoding strategy enforcing structure-preserving exploration.<n>We integrate this into Mixed-CUTS, a training framework synergizing exploitative and exploratory rollouts to amplify intra-group advantage variance.<n> Notably, Mixed-CUTS improves Pass@1 accuracy on the challenging AIME25 benchmark by up to 15.1% over standard GRPO.
arXiv Detail & Related papers (2026-04-20T16:43:28Z) - Robust Fine-tuning of Zero-shot Models via Variance Reduction [56.360865951192324]
When fine-tuning zero-shot models, our desideratum is for the fine-tuned model to excel in both in-distribution (ID) and out-of-distribution (OOD)
We propose a sample-wise ensembling technique that can simultaneously attain the best ID and OOD accuracy without the trade-offs.
arXiv Detail & Related papers (2024-11-11T13:13:39Z) - Trained Models Tell Us How to Make Them Robust to Spurious Correlation without Group Annotation [3.894771553698554]
Empirical Risk Minimization (ERM) models tend to rely on attributes that have high spurious correlation with the target.
This can degrade the performance on underrepresented (or'minority') groups that lack these attributes.
We propose Environment-based Validation and Loss-based Sampling (EVaLS) to enhance robustness to spurious correlation.
arXiv Detail & Related papers (2024-10-07T08:17:44Z) - PUMA: margin-based data pruning [51.12154122266251]
We focus on data pruning, where some training samples are removed based on the distance to the model classification boundary (i.e., margin)
We propose PUMA, a new data pruning strategy that computes the margin using DeepFool.
We show that PUMA can be used on top of the current state-of-the-art methodology in robustness, and it is able to significantly improve the model performance unlike the existing data pruning strategies.
arXiv Detail & Related papers (2024-05-10T08:02:20Z) - MixedNUTS: Training-Free Accuracy-Robustness Balance via Nonlinearly Mixed Classifiers [41.56951365163419]
"MixedNUTS" is a training-free method where the output logits of a robust classifier are processed by nonlinear transformations with only three parameters.
MixedNUTS then converts the transformed logits into probabilities and mixes them as the overall output.
On CIFAR-10, CIFAR-100, and ImageNet datasets, experimental results with custom strong adaptive attacks demonstrate MixedNUTS's vastly improved accuracy and near-SOTA robustness.
arXiv Detail & Related papers (2024-02-03T21:12:36Z) - Generalized Logit Adjustment: Calibrating Fine-tuned Models by Removing Label Bias in Foundation Models [75.9543301303586]
Foundation models like CLIP allow zero-shot transfer on various tasks without additional training data.
Fine-tuning and ensembling are also commonly adopted to better fit the downstream tasks.
However, we argue that prior work has overlooked the inherent biases in foundation models.
arXiv Detail & Related papers (2023-10-12T08:01:11Z) - Bias Amplification Enhances Minority Group Performance [10.380812738348899]
We propose BAM, a novel two-stage training algorithm.
In the first stage, the model is trained using a bias amplification scheme via introducing a learnable auxiliary variable for each training sample.
In the second stage, we upweight the samples that the bias-amplified model misclassifies, and then continue training the same model on the reweighted dataset.
arXiv Detail & Related papers (2023-09-13T04:40:08Z) - Ranking & Reweighting Improves Group Distributional Robustness [14.021069321266516]
We propose a ranking-based training method called Discounted Rank Upweighting (DRU) to learn models that exhibit strong OOD performance on the test data.
Results on several synthetic and real-world datasets highlight the superior ability of our group-ranking-based (akin to soft-minimax) approach in selecting and learning models that are robust to group distributional shifts.
arXiv Detail & Related papers (2023-05-09T20:37:16Z) - Distributionally Robust Models with Parametric Likelihood Ratios [123.05074253513935]
Three simple ideas allow us to train models with DRO using a broader class of parametric likelihood ratios.
We find that models trained with the resulting parametric adversaries are consistently more robust to subpopulation shifts when compared to other DRO approaches.
arXiv Detail & Related papers (2022-04-13T12:43:12Z) - Just Train Twice: Improving Group Robustness without Training Group
Information [101.84574184298006]
Standard training via empirical risk minimization can produce models that achieve high accuracy on average but low accuracy on certain groups.
Prior approaches that achieve high worst-group accuracy, like group distributionally robust optimization (group DRO) require expensive group annotations for each training point.
We propose a simple two-stage approach, JTT, that first trains a standard ERM model for several epochs, and then trains a second model that upweights the training examples that the first model misclassified.
arXiv Detail & Related papers (2021-07-19T17:52:32Z) - To be Robust or to be Fair: Towards Fairness in Adversarial Training [83.42241071662897]
We find that adversarial training algorithms tend to introduce severe disparity of accuracy and robustness between different groups of data.
We propose a Fair-Robust-Learning (FRL) framework to mitigate this unfairness problem when doing adversarial defenses.
arXiv Detail & Related papers (2020-10-13T02:21:54Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.