Abstract
Uncertainty approximation in text classification is an important area with applications in domain adaptation and interpretability. The most widely used uncertainty approximation method is Monte Carlo Dropout, which is computationally expensive as it requires multiple forward passes through the model. A cheaper alternative is to simply use a softmax to estimate model uncertainty. However, prior work has indicated that the softmax can generate overconfident uncertainty estimates and can thus be tricked into producing incorrect predictions. In this paper, we perform a thorough empirical analysis of both methods on five datasets with two base neural architectures in order to reveal insight into the trade-offs between the two. We compare the methods' uncertainty approximations and downstream text classification performance, while weighing their performance against their computational complexity as a cost-benefit analysis, by measuring runtime (cost) and the downstream performance (benefit). We find that, while Monte Carlo produces the best uncertainty approximations, using a simple softmax leads to competitive uncertainty estimation for text classification at a much lower computational cost, suggesting that softmax can in fact be a sufficient uncertainty estimate when computational resources are a concern.
Keyword: overconfidence
Adaptive Label Smoothing with Self-Knowledge in Natural Language Generation
Authors: Dongkyu Lee, Ka Chun Cheung, Nevin L. Zhang
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
Abstract
Overconfidence has been shown to impair generalization and calibration of a neural network. Previous studies remedy this issue by adding a regularization term to a loss function, preventing a model from making a peaked distribution. Label smoothing smoothes target labels with a pre-defined prior label distribution; as a result, a model is learned to maximize the likelihood of predicting the soft label. Nonetheless, the amount of smoothing is the same in all samples and remains fixed in training. In other words, label smoothing does not reflect the change in probability distribution mapped by a model over the course of training. To address this issue, we propose a regularization scheme that brings dynamic nature into the smoothing parameter by taking model probability distribution into account, thereby varying the parameter per instance. A model in training self-regulates the extent of smoothing on the fly during forward propagation. Furthermore, inspired by recent work in bridging label smoothing and knowledge distillation, our work utilizes self-knowledge as a prior label distribution in softening target labels, and presents theoretical support for the regularization effect by knowledge distillation and the dynamic smoothing parameter. Our regularizer is validated comprehensively, and the result illustrates marked improvements in model generalization and calibration, enhancing robustness and trustworthiness of a model.
Keyword: confidence
Adaptive Label Smoothing with Self-Knowledge in Natural Language Generation
Authors: Dongkyu Lee, Ka Chun Cheung, Nevin L. Zhang
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
Abstract
Overconfidence has been shown to impair generalization and calibration of a neural network. Previous studies remedy this issue by adding a regularization term to a loss function, preventing a model from making a peaked distribution. Label smoothing smoothes target labels with a pre-defined prior label distribution; as a result, a model is learned to maximize the likelihood of predicting the soft label. Nonetheless, the amount of smoothing is the same in all samples and remains fixed in training. In other words, label smoothing does not reflect the change in probability distribution mapped by a model over the course of training. To address this issue, we propose a regularization scheme that brings dynamic nature into the smoothing parameter by taking model probability distribution into account, thereby varying the parameter per instance. A model in training self-regulates the extent of smoothing on the fly during forward propagation. Furthermore, inspired by recent work in bridging label smoothing and knowledge distillation, our work utilizes self-knowledge as a prior label distribution in softening target labels, and presents theoretical support for the regularization effect by knowledge distillation and the dynamic smoothing parameter. Our regularizer is validated comprehensively, and the result illustrates marked improvements in model generalization and calibration, enhancing robustness and trustworthiness of a model.
On the Robustness of Dataset Inference
Authors: Sebastian Szyller, Rui Zhang, Jian Liu, N. Asokan
Subjects: Machine Learning (cs.LG); Cryptography and Security (cs.CR)
Abstract
Machine learning (ML) models are costly to train as they can require a significant amount of data, computational resources and technical expertise. Thus, they constitute valuable intellectual property that needs protection from adversaries wanting to steal them. Ownership verification techniques allow the victims of model stealing attacks to demonstrate that a suspect model was in fact stolen from theirs. Although a number of ownership verification techniques based on watermarking or fingerprinting have been proposed, most of them fall short either in terms of security guarantees (well-equipped adversaries can evade verification) or computational cost. A fingerprinting technique introduced at ICLR '21, Dataset Inference (DI), has been shown to offer better robustness and efficiency than prior methods. The authors of DI provided a correctness proof for linear (suspect) models. However, in the same setting, we prove that DI suffers from high false positives (FPs) -- it can incorrectly identify an independent model trained with non-overlapping data from the same distribution as stolen. We further prove that DI also triggers FPs in realistic, non-linear suspect models. We then confirm empirically that DI leads to FPs, with high confidence. Second, we show that DI also suffers from false negatives (FNs) -- an adversary can fool DI by regularising a stolen model's decision boundaries using adversarial training, thereby leading to an FN. To this end, we demonstrate that DI fails to identify a model adversarially trained from a stolen dataset -- the setting where DI is the hardest to evade. Finally, we discuss the implications of our findings, the viability of fingerprinting-based ownership verification in general, and suggest directions for future work.
Abstract
Question answering models can use rich knowledge sources -- up to one hundred retrieved passages and parametric knowledge in the large-scale language model (LM). Prior work assumes information in such knowledge sources is consistent with each other, paying little attention to how models blend information stored in their LM parameters with that from retrieved evidence documents. In this paper, we simulate knowledge conflicts (i.e., where parametric knowledge suggests one answer and different passages suggest different answers) and examine model behaviors. We find retrieval performance heavily impacts which sources models rely on, and current models mostly rely on non-parametric knowledge in their best-performing settings. We discover a troubling trend that contradictions among knowledge sources affect model confidence only marginally. To address this issue, we present a new calibration study, where models are discouraged from presenting any single answer when presented with multiple conflicting answer candidates in retrieved evidences.
Towards Robust Recommender Systems via Triple Cooperative Defense
Abstract
Recommender systems are often susceptible to well-crafted fake profiles, leading to biased recommendations. The wide application of recommender systems makes studying the defense against attack necessary. Among existing defense methods, data-processing-based methods inevitably exclude normal samples, while model-based methods struggle to enjoy both generalization and robustness. Considering the above limitations, we suggest integrating data processing and robust model and propose a general framework, Triple Cooperative Defense (TCD), which cooperates to improve model robustness through the co-training of three models. Specifically, in each round of training, we sequentially use the high-confidence prediction ratings (consistent ratings) of any two models as auxiliary training data for the remaining model, and the three models cooperatively improve recommendation robustness. Notably, TCD adds pseudo label data instead of deleting abnormal data, which avoids the cleaning of normal data, and the cooperative training of the three models is also beneficial to model generalization. Through extensive experiments with five poisoning attacks on three real-world datasets, the results show that the robustness improvement of TCD significantly outperforms baselines. It is worth mentioning that TCD is also beneficial for model generalizations.
Confidence-Calibrated Face and Kinship Verification
Authors: Min Xu, Ximiao Zhang, Xiuzhuang Zhou
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
In this paper, we investigate the problem of predictive confidence in face and kinship verification. Most existing face and kinship verification methods focus on accuracy performance while ignoring confidence estimation for their prediction results. However, confidence estimation is essential for modeling reliability in such high-risk tasks. To address this issue, we first introduce a novel yet simple confidence measure for face and kinship verification, which allows the verification models to transform the similarity score into a confidence score for a given face pair. We further propose a confidence-calibrated approach called angular scaling calibration (ASC). ASC is easy to implement and can be directly applied to existing face and kinship verification models without model modifications, yielding accuracy-preserving and confidence-calibrated probabilistic verification models. To the best of our knowledge, our approach is the first general confidence-calibrated solution to face and kinship verification in a modern context. We conduct extensive experiments on four widely used face and kinship verification datasets, and the results demonstrate the effectiveness of our approach.
Abstract
An important component in deploying machine learning (ML) in safety-critic applications is having a reliable measure of confidence in the ML model's predictions. For a classifier $f$ producing a probability vector $f(x)$ over the candidate classes, the confidence is typically taken to be $\max_i f(x)_i$. This approach is potentially limited, as it disregards the rest of the probability vector. In this work, we derive several confidence measures that depend on information beyond the maximum score, such as margin-based and entropy-based measures, and empirically evaluate their usefulness, focusing on NLP tasks with distribution shifts and Transformer-based models. We show that when models are evaluated on the out-of-distribution data ``out of the box'', using only the maximum score to inform the confidence measure is highly suboptimal. In the post-processing regime (where the scores of $f$ can be improved using additional in-distribution held-out data), this remains true, albeit less significant. Overall, our results suggest that entropy-based confidence is a surprisingly useful measure.
Keyword: scaling
Scaling up and Stabilizing Differentiable Planning with Implicit Differentiation
Authors: Linfeng Zhao, Huazhe Xu, Lawson L.S. Wong
Abstract
Differentiable planning promises end-to-end differentiability and adaptivity. However, an issue prevents it from scaling up to larger-scale problems: they need to differentiate through forward iteration layers to compute gradients, which couples forward computation and backpropagation, and needs to balance forward planner performance and computational cost of the backward pass. To alleviate this issue, we propose to differentiate through the Bellman fixed-point equation to decouple forward and backward passes for Value Iteration Network and its variants, which enables constant backward cost (in planning horizon) and flexible forward budget and helps scale up to large tasks. We study the convergence stability, scalability, and efficiency of the proposed implicit version of VIN and its variants and demonstrate their superiorities on a range of planning tasks: 2D navigation, visual navigation, and 2-DOF manipulation in configuration space and workspace.
Confidence-Calibrated Face and Kinship Verification
Authors: Min Xu, Ximiao Zhang, Xiuzhuang Zhou
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
In this paper, we investigate the problem of predictive confidence in face and kinship verification. Most existing face and kinship verification methods focus on accuracy performance while ignoring confidence estimation for their prediction results. However, confidence estimation is essential for modeling reliability in such high-risk tasks. To address this issue, we first introduce a novel yet simple confidence measure for face and kinship verification, which allows the verification models to transform the similarity score into a confidence score for a given face pair. We further propose a confidence-calibrated approach called angular scaling calibration (ASC). ASC is easy to implement and can be directly applied to existing face and kinship verification models without model modifications, yielding accuracy-preserving and confidence-calibrated probabilistic verification models. To the best of our knowledge, our approach is the first general confidence-calibrated solution to face and kinship verification in a modern context. We conduct extensive experiments on four widely used face and kinship verification datasets, and the results demonstrate the effectiveness of our approach.
From colouring-in to pointillism: revisiting semantic segmentation supervision
Authors: Rodrigo Benenson, Vittorio Ferrari
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
The prevailing paradigm for producing semantic segmentation training data relies on densely labelling each pixel of each image in the training set, akin to colouring-in books. This approach becomes a bottleneck when scaling up in the number of images, classes, and annotators. Here we propose instead a pointillist approach for semantic segmentation annotation, where only point-wise yes/no questions are answered. We explore design alternatives for such an active learning approach, measure the speed and consistency of human annotators on this task, show that this strategy enables training good segmentation models, and that it is suitable for evaluating models at test time. As concrete proof of the scalability of our method, we collected and released 22.6M point labels over 4,171 classes on the Open Images dataset. Our results enable to rethink the semantic segmentation pipeline of annotation, training, and evaluation from a pointillism point of view.
Practical Program Repair in the Era of Large Pre-trained Language Models
Authors: Chunqiu Steven Xia, Yuxiang Wei, Lingming Zhang
Abstract
Automated Program Repair (APR) aims to help developers automatically patch software bugs. However, current state-of-the-art traditional and learning-based APR techniques face the problem of limited patch variety, failing to fix complicated bugs. This is mainly due to the reliance on bug-fixing datasets to craft fix templates or directly predict potential patches. Large Pre-Trained Language Models (PLMs), trained using billions of text/code tokens, can potentially help avoid this issue. Very recently, researchers have directly leveraged PLMs for APR without relying on any bug-fixing datasets. Meanwhile, such existing work either failed to include state-of-the-art PLMs or was not evaluated on realistic datasets. In this work, we perform the first extensive study on directly applying PLMs for APR. We select 9 recent state-of-the-art PLMs, including both generative and infilling models, ranging from 125M to 20B in size. We designed 3 different repair settings to evaluate the different ways we can use PLMs to generate patches. We apply the PLMs under these repair settings on 5 datasets across 3 different languages and compare different PLMs in the number of bugs fixed, generation speed and compilation rate. Our study demonstrates that directly applying state-of-the-art PLMs can already substantially outperform all existing APR techniques on all our datasets. Among the studied PLMs, the scaling effect exists for APR where larger models tend to achieve better performance. Also, we show for the first time that suffix code after the buggy line (adopted in infilling-style APR) is important in not only generating more fixes but more patches with higher compilation rate. Besides patch generation, the PLMs consider correct patches to be more natural than other ones, and can even be leveraged for effective patch ranking or patch correctness checking.
Keyword: calibration
Adaptive Label Smoothing with Self-Knowledge in Natural Language Generation
Authors: Dongkyu Lee, Ka Chun Cheung, Nevin L. Zhang
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
Abstract
Overconfidence has been shown to impair generalization and calibration of a neural network. Previous studies remedy this issue by adding a regularization term to a loss function, preventing a model from making a peaked distribution. Label smoothing smoothes target labels with a pre-defined prior label distribution; as a result, a model is learned to maximize the likelihood of predicting the soft label. Nonetheless, the amount of smoothing is the same in all samples and remains fixed in training. In other words, label smoothing does not reflect the change in probability distribution mapped by a model over the course of training. To address this issue, we propose a regularization scheme that brings dynamic nature into the smoothing parameter by taking model probability distribution into account, thereby varying the parameter per instance. A model in training self-regulates the extent of smoothing on the fly during forward propagation. Furthermore, inspired by recent work in bridging label smoothing and knowledge distillation, our work utilizes self-knowledge as a prior label distribution in softening target labels, and presents theoretical support for the regularization effect by knowledge distillation and the dynamic smoothing parameter. Our regularizer is validated comprehensively, and the result illustrates marked improvements in model generalization and calibration, enhancing robustness and trustworthiness of a model.
Abstract
Question answering models can use rich knowledge sources -- up to one hundred retrieved passages and parametric knowledge in the large-scale language model (LM). Prior work assumes information in such knowledge sources is consistent with each other, paying little attention to how models blend information stored in their LM parameters with that from retrieved evidence documents. In this paper, we simulate knowledge conflicts (i.e., where parametric knowledge suggests one answer and different passages suggest different answers) and examine model behaviors. We find retrieval performance heavily impacts which sources models rely on, and current models mostly rely on non-parametric knowledge in their best-performing settings. We discover a troubling trend that contradictions among knowledge sources affect model confidence only marginally. To address this issue, we present a new calibration study, where models are discouraged from presenting any single answer when presented with multiple conflicting answer candidates in retrieved evidences.
Parameter-Efficient Legal Domain Adaptation
Authors: Jonathan Li, Rohan Bhambhoria, Xiaodan Zhu
Abstract
Seeking legal advice is often expensive. Recent advancement in machine learning for solving complex problems can be leveraged to help make legal services more accessible to the public. However, real-life applications encounter significant challenges. State-of-the-art language models are growing increasingly large, making parameter-efficient learning increasingly important. Unfortunately, parameter-efficient methods perform poorly with small amounts of data, which are common in the legal domain (where data labelling costs are high). To address these challenges, we propose parameter-efficient legal domain adaptation, which uses vast unsupervised legal data from public legal forums to perform legal pre-training. This method exceeds or matches the fewshot performance of existing models such as LEGAL-BERT on various legal tasks while tuning only approximately 0.1% of model parameters. Additionally, we show that our method can achieve calibration comparable to existing methods across several tasks. To the best of our knowledge, this work is among the first to explore parameter-efficient methods of tuning language models toward the legal domain.
Confidence-Calibrated Face and Kinship Verification
Authors: Min Xu, Ximiao Zhang, Xiuzhuang Zhou
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
In this paper, we investigate the problem of predictive confidence in face and kinship verification. Most existing face and kinship verification methods focus on accuracy performance while ignoring confidence estimation for their prediction results. However, confidence estimation is essential for modeling reliability in such high-risk tasks. To address this issue, we first introduce a novel yet simple confidence measure for face and kinship verification, which allows the verification models to transform the similarity score into a confidence score for a given face pair. We further propose a confidence-calibrated approach called angular scaling calibration (ASC). ASC is easy to implement and can be directly applied to existing face and kinship verification models without model modifications, yielding accuracy-preserving and confidence-calibrated probabilistic verification models. To the best of our knowledge, our approach is the first general confidence-calibrated solution to face and kinship verification in a modern context. We conduct extensive experiments on four widely used face and kinship verification datasets, and the results demonstrate the effectiveness of our approach.
Keyword: out of distribution detection
There is no result
Keyword: out-of-distribution detection
There is no result
Keyword: expected calibration error
There is no result
Keyword: overconfident
Revisiting Softmax for Uncertainty Approximation in Text Classification
Keyword: overconfidence
Adaptive Label Smoothing with Self-Knowledge in Natural Language Generation
Keyword: confidence
Adaptive Label Smoothing with Self-Knowledge in Natural Language Generation
On the Robustness of Dataset Inference
Rich Knowledge Sources Bring Complex Knowledge Conflicts: Recalibrating Models to Reflect Conflicting Evidence
Towards Robust Recommender Systems via Triple Cooperative Defense
Confidence-Calibrated Face and Kinship Verification
Useful Confidence Measures: Beyond the Max Score
Keyword: scaling
Scaling up and Stabilizing Differentiable Planning with Implicit Differentiation
Confidence-Calibrated Face and Kinship Verification
From colouring-in to pointillism: revisiting semantic segmentation supervision
Practical Program Repair in the Era of Large Pre-trained Language Models
Keyword: calibration
Adaptive Label Smoothing with Self-Knowledge in Natural Language Generation
Rich Knowledge Sources Bring Complex Knowledge Conflicts: Recalibrating Models to Reflect Conflicting Evidence
Parameter-Efficient Legal Domain Adaptation
Confidence-Calibrated Face and Kinship Verification