• Corpus ID: 244709204

Confounder Identification-free Causal Visual Feature Learning

@article{Li2021ConfounderIC,
  title={Confounder Identification-free Causal Visual Feature Learning},
  author={Xin Li and Zhizheng Zhang and Guoqiang Wei and Cuiling Lan and Wenjun Zeng and Xin Jin and Zhibo Chen},
  journal={ArXiv},
  year={2021},
  volume={abs/2111.13420}
}
Confounders in deep learning are in general detrimental to model’s generalization where they infiltrate feature representations. Therefore, learning causal features that are free of interference from confounders is important. Most previous causal learning based approaches employ back-door criterion to mitigate the adverse effect of certain specific confounder, which require the explicit identification of confounder. However, in real scenarios, confounders are typically diverse and difficult to… 
Causal Reasoning Meets Visual Representation Learning: A Prospective Study
TLDR
This paper conducts a comprehensive review of existing causal reasoning methods for visual representation learning, covering fundamental theories, models, and datasets, and proposes some prospective challenges, opportunities, and future research directions for benchmarking causal reasoning algorithms inVisual representation learning.
Deep Frequency Filtering for Domain Generalization
TLDR
Extensive experiments demonstrate the effectiveness of the proposed Deep Frequency Filtering and show that applying DFF on a plain baseline outperforms the state-of-the-art methods on different domain generalization tasks, including close-set classification and open-set retrieval.
Causal Reasoning with Spatial-temporal Representation Learning: A Prospective Study
TLDR
This paper conducts a comprehensive review of existing causal reasoning methods for spatial-temporal representation learning, covering fundamental theories, models, and datasets, and proposes some primary challenges, opportunities, and future research directions for benchmarking causal reasoning algorithms in spatial- temporal representation learning.

References

SHOWING 1-10 OF 71 REFERENCES
Causal Attention for Vision-Language Tasks
TLDR
CATT improves various popular attention-based vision-language models by considerable margins and has great potential in large-scale pre-training, e.g., it can promote the lighter LXMERT, which uses fewer data and less computational power, comparable to the heavier UNITER.
Interventional Few-Shot Learning
TLDR
It is revealed that the contribution of IFSL is orthogonal to existing fine-tuning and meta-learning based FSL methods, hence IFSL can improve all of them, achieving a new 1-/5-shot state-of-the-art on \textit{mini}ImageNet, \text it{tiered}Image net, and cross-domain CUB.
Transporting Causal Mechanisms for Unsupervised Domain Adaptation
TLDR
Transporting Causal Mechanisms (TCM) is proposed, to identify the confounder stratum and representations by using the domain-invariant disentangled causal mechanisms, which are discovered in an unsupervised fashion.
Distilling Causal Effect of Data in Class-Incremental Learning
TLDR
The proposed causal effect distillation method can improve various state-of-the-art CIL methods by a large margin and capture the Incremental Momentum Effect of the data stream, which can help to retain the old effect overwhelmed by the new data effect, and thus alleviate the forgetting of the old class in testing.
Visual Commonsense Representation Learning via Causal Inference
We present a novel unsupervised feature representation learning method, Visual Commonsense Region-based Convolutional Neural Network (VC R-CNN1), to serve as an improved visual region encoder for
Domain Generalization using Causal Matching
TLDR
An iterative algorithm called MatchDG is proposed that approximates base object similarity by using a contrastive loss formulation adapted for multiple domains and learns matches that have over 25\% overlap with ground-truth object matches in MNIST and Fashion-MNIST.
Causal Intervention for Weakly-Supervised Semantic Segmentation
TLDR
A structural causal model to analyze the causalities among images, contexts, and class labels is proposed and a new method: Context Adjustment (CONTA) is developed, to remove the confounding bias in image-level classification and thus provide better pseudo-masks as ground-truth for the subsequent segmentation model.
Learning to Generalize Unseen Domains via Memory-based Multi-Source Meta-Learning for Person Re-Identification
TLDR
The Memory-based Multi-Source Meta-Learning (M3L) framework to train a generalizable model for unseen domains is proposed and a meta-learning strategy is introduced to simulate the train-test process of domain generalization for learning more generalizable models.
Domain Generalization via Model-Agnostic Learning of Semantic Features
TLDR
This work investigates the challenging problem of domain generalization, i.e., training a model on multi-domain source data such that it can directly generalize to target domains with unknown statistics, and adopts a model-agnostic learning paradigm with gradient-based meta-train and meta-test procedures to expose the optimization to domain shift.
Domain Generalization by Solving Jigsaw Puzzles
TLDR
This model learns the semantic labels in a supervised fashion, and broadens its understanding of the data by learning from self-supervised signals how to solve a jigsaw puzzle on the same images, which helps the network to learn the concepts of spatial correlation while acting as a regularizer for the classification task.
...
...