A multi-label classifier estimates the binary label state (relevant vs irrelevant) for each of a set of concept labels, for any given instance. Probabilistic multi-label classifiers provide a predictive posterior distribution over all possible labelset combinations of such label states (the powerset of labels) from which we can provide the best estimate, simply by selecting the labelset corresponding to the largest expected accuracy, over that distribution. For example, in maximizing exact match accuracy, we provide the mode of the distribution. But how does this relate to the confidence we may have in such an estimate? Confidence is an important element of real-world applications of multi-label classifiers (as in machine learning in general) and is an important ingredient in explainability and interpretability. However, it is not obvious how to provide confidence in the multi-label context and relating to a particular accuracy metric, and nor is it clear how to provide a confidence which correlates well with the expected accuracy, which would be most valuable in real-world decision making. In this article we estimate the expected accuracy as a surrogate for confidence, for a given accuracy metric. We hypothesise that the expected accuracy can be estimated from the multi-label predictive distribution. We examine seven candidate functions for their ability to estimate expected accuracy from the predictive distribution. We found three of these to correlate to expected accuracy and are robust. Further, we determined that each candidate function can be used separately to estimate Hamming similarity, but a combination of the candidates was best for expected Jaccard index and exact match.
相關內容
Deep neural networks have achieved impressive performance in a variety of tasks over the last decade, such as autonomous driving, face recognition, and medical diagnosis. However, prior works show that deep neural networks are easily manipulated into specific, attacker-decided behaviors in the inference stage by backdoor attacks which inject malicious small hidden triggers into model training, raising serious security threats. To determine the triggered neurons and protect against backdoor attacks, we exploit Shapley value and develop a new approach called Shapley Pruning (ShapPruning) that successfully mitigates backdoor attacks from models in a data-insufficient situation (1 image per class or even free of data). Considering the interaction between neurons, ShapPruning identifies the few infected neurons (under 1% of all neurons) and manages to protect the model's structure and accuracy after pruning as many infected neurons as possible. To accelerate ShapPruning, we further propose discarding threshold and $\epsilon$-greedy strategy to accelerate Shapley estimation, making it possible to repair poisoned models with only several minutes. Experiments demonstrate the effectiveness and robustness of our method against various attacks and tasks compared to existing methods.
Collecting complete network data is expensive, time-consuming, and often infeasible. Aggregated Relational Data (ARD), which capture information about a social network by asking a respondent questions of the form ``How many people with trait X do you know?'' provide a low-cost option when collecting complete network data is not possible. Rather than asking about connections between each pair of individuals directly, ARD collects the number of contacts the respondent knows with a given trait. Despite widespread use and a growing literature on ARD methodology, there is still no systematic understanding of when and why ARD should accurately recover features of the unobserved network. This paper provides such a characterization by deriving conditions under which statistics about the unobserved network (or functions of these statistics like regression coefficients) can be consistently estimated using ARD. We do this by first providing consistent estimates of network model parameters for three commonly used probabilistic models: the beta-model with node-specific unobserved effects, the stochastic block model with unobserved community structure, and latent geometric space models with unobserved latent locations. A key observation behind these results is that cross-group link probabilities for a collection of (possibly unobserved) groups identifies the model parameters, meaning ARD is sufficient for parameter estimation. With these estimated parameters, it is possible to simulate graphs from the fitted distribution and analyze the distribution of network statistics. We can then characterize conditions under which the simulated networks based on ARD will allow for consistent estimation of the unobserved network statistics, such as eigenvector centrality or response functions by or of the unobserved network, such as regression coefficients.
6D pose estimation of textureless objects is a valuable but challenging task for many robotic applications. In this work, we propose a framework to address this challenge using only RGB images acquired from multiple viewpoints. The core idea of our approach is to decouple 6D pose estimation into a sequential two-step process, first estimating the 3D translation and then the 3D rotation of each object. This decoupled formulation first resolves the scale and depth ambiguities in single RGB images, and uses these estimates to accurately identify the object orientation in the second stage, which is greatly simplified with an accurate scale estimate. Moreover, to accommodate the multi-modal distribution present in rotation space, we develop an optimization scheme that explicitly handles object symmetries and counteracts measurement uncertainties. In comparison to the state-of-the-art multi-view approach, we demonstrate that the proposed approach achieves substantial improvements on a challenging 6D pose estimation dataset for textureless objects.
Generative data-free quantization emerges as a practical compression approach that quantizes deep neural networks to low bit-width without accessing the real data. This approach generates data utilizing batch normalization (BN) statistics of the full-precision networks to quantize the networks. However, it always faces the serious challenges of accuracy degradation in practice. We first give a theoretical analysis that the diversity of synthetic samples is crucial for the data-free quantization, while in existing approaches, the synthetic data completely constrained by BN statistics experimentally exhibit severe homogenization at distribution and sample levels. This paper presents a generic Diverse Sample Generation (DSG) scheme for the generative data-free quantization, to mitigate detrimental homogenization. We first slack the statistics alignment for features in the BN layer to relax the distribution constraint. Then, we strengthen the loss impact of the specific BN layers for different samples and inhibit the correlation among samples in the generation process, to diversify samples from the statistical and spatial perspectives, respectively. Comprehensive experiments show that for large-scale image classification tasks, our DSG can consistently quantization performance on different neural architectures, especially under ultra-low bit-width. And data diversification caused by our DSG brings a general gain to various quantization-aware training and post-training quantization approaches, demonstrating its generality and effectiveness.
Pre-trained language models (PrLMs) have achieved great success on a wide range of natural language processing tasks by virtue of the universal language representation ability obtained by self-supervised learning on a large corpus. These models are pre-trained on standard plain texts with general language model (LM) training objectives, which would be insufficient to model dialogue-exclusive attributes like specificity and informativeness reflected in these tasks that are not explicitly captured by the pre-trained universal language representations. In this work, we propose dialogue-adaptive pre-training objectives (DAPO) derived from quality estimation to simulate dialogue-specific features, namely coherence, specificity, and informativeness. As the foundation for model pre-training, we synthesize a new dialogue corpus and build our training set with two unsupervised methods: 1) coherence-oriented context corruption, including utterance ordering, insertion, and replacement, to help the model capture the coherence inside the dialogue contexts; and 2) specificity-oriented automatic rescoring, which encourages the model to measure the quality of the synthesized data for dialogue-adaptive pre-training by considering specificity and informativeness. Experimental results on widely used open-domain response selection and quality estimation benchmarks show that DAPO significantly improves the baseline models and achieves state-of-the-art performance on the MuTual leaderboard, verifying the effectiveness of estimating quality evaluation factors into pre-training.
Volatility clustering is a common phenomenon in financial time series. Typically, linear models can be used to describe the temporal autocorrelation of the (logarithmic) variance of returns. Considering the difficulty in estimating this model, we construct a Dynamic Bayesian Network, which utilizes the conjugate prior relation of normal-gamma and gamma-gamma, so that its posterior form locally remains unchanged at each node. This makes it possible to find approximate solutions using variational methods quickly. Furthermore, we ensure that the volatility expressed by the model is an independent incremental process after inserting dummy gamma nodes between adjacent time steps. We have found that this model has two advantages: 1) It can be proved that it can express heavier tails than Gaussians, i.e., have positive excess kurtosis, compared to popular linear models. 2) If the variational inference(VI) is used for state estimation, it runs much faster than Monte Carlo(MC) methods since the calculation of the posterior uses only basic arithmetic operations. And its convergence process is deterministic. We tested the model, named Gam-Chain, using recent Crypto, Nasdaq, and Forex records of varying resolutions. The results show that: 1) In the same case of using MC, this model can achieve comparable state estimation results with the regular lognormal chain. 2) In the case of only using VI, this model can obtain accuracy that are slightly worse than MC, but still acceptable in practice; 3) Only using VI, the running time of Gam-Chain, in general case, can be reduced to below 5% of that based on the lognormal chain via MC.
Anomaly detection methods identify examples that do not follow the expected behaviour, typically in an unsupervised fashion, by assigning real-valued anomaly scores to the examples based on various heuristics. These scores need to be transformed into actual predictions by thresholding, so that the proportion of examples marked as anomalies equals the expected proportion of anomalies, called contamination factor. Unfortunately, there are no good methods for estimating the contamination factor itself. We address this need from a Bayesian perspective, introducing a method for estimating the posterior distribution of the contamination factor of a given unlabeled dataset. We leverage on outputs of several anomaly detectors as a representation that already captures the basic notion of anomalousness and estimate the contamination using a specific mixture formulation. Empirically on 22 datasets, we show that the estimated distribution is well-calibrated and that setting the threshold using the posterior mean improves the anomaly detectors' performance over several alternative methods. All code is publicly available for full reproducibility.
In this paper we integrate the isotonic regression with Stone's cross-validation-based method to estimate discrete infinitely supported distribution. We prove that the estimator is strongly consistent, derive its rate of convergence for any underlying distribution, and for one-dimensional case we derive Marshal-type inequality for cumulative distribution function of the estimator. Also, we construct the asymptotically correct conservative global confidence band for the estimator. It is shown that, first, the estimator performs good even for small sized data sets, second, the estimator outperforms in the case of non-monotone underlying distribution, and, third, it performs almost as good as Grenander estimator when the true distribution is isotonic. Therefore, the new estimator provides a trade-off between goodness-of-fit, monotonicity and quality of probabilistic forecast. We apply the estimator to the time-to-onset data of visceral leishmaniasis in Brazil collected from 2007 to 2014.
With advancements in computer vision taking place day by day, recently a lot of light is being shed on activity recognition. With the range for real-world applications utilizing this field of study increasing across a multitude of industries such as security and healthcare, it becomes crucial for businesses to distinguish which machine learning methods perform better than others in the area. This paper strives to aid in this predicament i.e. building upon previous related work, it employs both classical and ensemble approaches on rich pose estimation (OpenPose) and HAR datasets. Making use of appropriate metrics to evaluate the performance for each model, the results show that overall, random forest yields the highest accuracy in classifying ADLs. Relatively all the models have excellent performance across both datasets, except for logistic regression and AdaBoost perform poorly in the HAR one. With the limitations of this paper also discussed in the end, the scope for further research is vast, which can use this paper as a base in aims of producing better results.
In this paper, we study the few-shot multi-label classification for user intent detection. For multi-label intent detection, state-of-the-art work estimates label-instance relevance scores and uses a threshold to select multiple associated intent labels. To determine appropriate thresholds with only a few examples, we first learn universal thresholding experience on data-rich domains, and then adapt the thresholds to certain few-shot domains with a calibration based on nonparametric learning. For better calculation of label-instance relevance score, we introduce label name embedding as anchor points in representation space, which refines representations of different classes to be well-separated from each other. Experiments on two datasets show that the proposed model significantly outperforms strong baselines in both one-shot and five-shot settings.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191