AI-KD: Adversarial learning and Implicit regularization for self-Knowledge Distillation

• URL: http://arxiv.org/abs/2211.10938v2
• Date: Thu, 21 Mar 2024 13:51:10 GMT
• Title: AI-KD: Adversarial learning and Implicit regularization for self-Knowledge Distillation
• Authors: Hyungmin Kim, Sungho Suh, Sunghyun Baek, Daehwan Kim, Daun Jeong, Hansang Cho, Junmo Kim,
• Abstract summary: We present a novel adversarial penalized self-knowledge distillation method, named adversarial learning and implicit regularization for self-knowledge distillation (AI-KD) Our model not only distills the deterministic and progressive knowledge which are from the pre-trained and previous epoch predictive probabilities but also transfers the knowledge of the deterministic predictive distributions using adversarial learning. We demonstrate the effectiveness of the proposed method with network architectures on multiple datasets and show the proposed method achieves better performance than state-of-the-art methods.
• Score: 20.80341206396388
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We present a novel adversarial penalized self-knowledge distillation method, named adversarial learning and implicit regularization for self-knowledge distillation (AI-KD), which regularizes the training procedure by adversarial learning and implicit distillations. Our model not only distills the deterministic and progressive knowledge which are from the pre-trained and previous epoch predictive probabilities but also transfers the knowledge of the deterministic predictive distributions using adversarial learning. The motivation is that the self-knowledge distillation methods regularize the predictive probabilities with soft targets, but the exact distributions may be hard to predict. Our method deploys a discriminator to distinguish the distributions between the pre-trained and student models while the student model is trained to fool the discriminator in the trained procedure. Thus, the student model not only can learn the pre-trained model's predictive probabilities but also align the distributions between the pre-trained and student models. We demonstrate the effectiveness of the proposed method with network architectures on multiple datasets and show the proposed method achieves better performance than state-of-the-art methods.

Related papers

• Sparse Logit Sampling: Accelerating Knowledge Distillation in LLMs [12.73155638335145]
  We prove that naive approaches for sparse knowledge distillation such as caching Top-K probabilities, while intuitive, provide biased estimates of teacher probability distribution to the student. We propose an importance-sampling-based method Random Sampling Knowledge Distillation', which provides unbiased estimates, preserves the gradient in expectation, and requires storing significantly sparser logits.
  arXiv  Detail & Related papers  (2025-03-21T05:58:18Z)
• Self-Knowledge Distillation for Learning Ambiguity [11.755814660833549]
  Recent language models often over-confidently predict a single label without consideration for its correctness. We propose a novel self-knowledge distillation method that enables models to learn label distributions more accurately. We validate our method on diverse NLU benchmark datasets and the experimental results demonstrate its effectiveness in producing better label distributions.
  arXiv  Detail & Related papers  (2024-06-14T05:11:32Z)
• Selective Learning: Towards Robust Calibration with Dynamic Regularization [79.92633587914659]
  Miscalibration in deep learning refers to there is a discrepancy between the predicted confidence and performance. We introduce Dynamic Regularization (DReg) which aims to learn what should be learned during training thereby circumventing the confidence adjusting trade-off.
  arXiv  Detail & Related papers  (2024-02-13T11:25:20Z)
• Self-Knowledge Distillation via Dropout [0.7883397954991659]
  We propose a simple and effective self-knowledge distillation method using a dropout (SD-Dropout) Our method does not require any additional trainable modules, does not rely on data, and requires only simple operations.
  arXiv  Detail & Related papers  (2022-08-11T05:08:55Z)
• Agree to Disagree: Diversity through Disagreement for Better Transferability [54.308327969778155]
  We propose D-BAT (Diversity-By-disAgreement Training), which enforces agreement among the models on the training data. We show how D-BAT naturally emerges from the notion of generalized discrepancy.
  arXiv  Detail & Related papers  (2022-02-09T12:03:02Z)
• Leveraging Unlabeled Data to Predict Out-of-Distribution Performance [63.740181251997306]
  Real-world machine learning deployments are characterized by mismatches between the source (training) and target (test) distributions. In this work, we investigate methods for predicting the target domain accuracy using only labeled source data and unlabeled target data. We propose Average Thresholded Confidence (ATC), a practical method that learns a threshold on the model's confidence, predicting accuracy as the fraction of unlabeled examples.
  arXiv  Detail & Related papers  (2022-01-11T23:01:12Z)
• Contrastive Learning for Fair Representations [50.95604482330149]
  Trained classification models can unintentionally lead to biased representations and predictions. Existing debiasing methods for classification models, such as adversarial training, are often expensive to train and difficult to optimise. We propose a method for mitigating bias by incorporating contrastive learning, in which instances sharing the same class label are encouraged to have similar representations.
  arXiv  Detail & Related papers  (2021-09-22T10:47:51Z)
• A Simple Framework for Uncertainty in Contrastive Learning [11.64841553345271]
  We introduce a simple approach that learns to assign uncertainty for pretrained contrastive representations. We train a deep network from a representation to a distribution in representation space, whose variance can be used as a measure of confidence. In our experiments, we show that this deep uncertainty model can be used (1) to visually interpret model behavior, (2) to detect new noise in the input to deployed models, (3) to detect anomalies, where we outperform 10 baseline methods across 11 tasks with improvements of up to 14% absolute.
  arXiv  Detail & Related papers  (2020-10-05T14:17:42Z)
• Learning Diverse Representations for Fast Adaptation to Distribution Shift [78.83747601814669]
  We present a method for learning multiple models, incorporating an objective that pressures each to learn a distinct way to solve the task. We demonstrate our framework's ability to facilitate rapid adaptation to distribution shift.
  arXiv  Detail & Related papers  (2020-06-12T12:23:50Z)
• Incremental Learning for End-to-End Automatic Speech Recognition [41.297106772785206]
  We propose an incremental learning method for end-to-end Automatic Speech Recognition (ASR) We design a novel explainability-based knowledge distillation for ASR models, which is combined with a response-based knowledge distillation to maintain the original model's predictions and the "reason" for the predictions. Results on a multi-stage sequential training task show that our method outperforms existing ones in mitigating forgetting.
  arXiv  Detail & Related papers  (2020-05-11T08:18:08Z)
• Binary Classification from Positive Data with Skewed Confidence [85.18941440826309]
  Positive-confidence (Pconf) classification is a promising weakly-supervised learning method. In practice, the confidence may be skewed by bias arising in an annotation process. We introduce the parameterized model of the skewed confidence, and propose the method for selecting the hyper parameter.
  arXiv  Detail & Related papers  (2020-01-29T00:04:36Z)

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.