Deep Similarity-Based Batch Mode Active Learning with Exploration-Exploitation

Yin, Changchang; Qian, Buyue; Cao, Shilei; Li, Xiaoyü; Wei, Jishang; Davidson, Ian

doi:10.1109/icdm.2017.67

Cited by 32 publications

(20 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In dataset R1a, we use 11.89% of training data to achieve 85.48% test accuracy compared to 87.24% in fully supervised training. Moreover, we compared with 8 representative active learning baselines, including methods using a single query principle, namely, Random Query (Woo and Park, 2012), Uncertainty Query (Joshi et al, 2009), CoreSet Query (Sener and Savarese, 2018), Bayesian Query (Gal et al, 2017), Bayesian Generative Active Learning (Tran et al, 2019) (BGAL) and hybrid query heuristics, exploration-exploitation BMAL (Yin et al, 2017) (EE-BMAL), VAAL (Sinha et al, 2019) and BADGE (Ash et al, 2020). As shown in Tab.1, our method achieves a superior performance on all 5 different datasets under the same labeling budget.…”

Section: Results and Comparisonsmentioning

confidence: 99%

“…Some hybrid query approaches are also proposed. For instance, Yin et al (2017) selected the data by uncertainty sampling and random sampling while we use a discriminator to query representative samples in order to reduce sampling bias in uncertainty sampling. Ash et al (2020) selected samples whose gradients span a diverse directions, which did not explicitly consider different query heuristics.…”

Section: Active Learningmentioning

confidence: 99%

“…The overall learning and querying steps are summarized in Figure 2. Note that the hybrid querying heuristics are also proposed in literature (Yin et al, 2017;Ash et al, 2020) and we defer the discussion in Section 2.2. The contributions of this paper is summarized as follows: 1) We propose a 3D HAL framework to query unlabeled subtomogram samples and expand the training dataset, such that deep models can be trained with significantly lower labeling cost while incurring minimal prediction accuracy drop.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Active Learning to Classify Macromolecular Structures in situ for Less Supervision in Cryo-Electron Tomography

Du,

Wang,

Zhu

et al. 2021

Preprint

View full text Add to dashboard Cite

Motivation: Cryo-Electron Tomography (cryo-ET) is a 3D bioimaging tool that visualizes the structural and spatial organization of macromolecules at a near-native state in single cells, which has broad applications in life science. However, the systematic structural recognition and recovery of macromolecules captured by cryo-ET are difficult due to high structural complexity and imaging limits. Deep learning based subtomogram classification have played critical roles for such tasks. As supervised approaches, however, their performance relies on sufficient and laborious annotation on a large training dataset. Results: To alleviate this major labeling burden, we proposed a Hybrid Active Learning (HAL) framework for querying subtomograms for labelling from a large unlabeled subtomogram pool. Firstly, HAL adopts uncertainty sampling to select the subtomograms that have the most uncertain predictions. This strategy enforces the model to be aware of the inductive bias during classification and subtomogram selection, which satisfies the discriminativeness principle in AL literature. Moreover, to mitigate the sampling bias caused by such strategy, a discriminator is introduced to judge if a certain subtomogram is labeled or unlabeled and subsequently the model queries the subtomogram that have higher probabilities to be unlabeled. Such query strategy encourages to match the data distribution between the labeled and unlabeled subtomogram samples, which essentially encodes the representativeness criterion into the subtomogram selection process. Additionally, HAL introduces a subset sampling strategy to improve the diversity of the query set, so that the information overlap is decreased between the queried batches and the algorithmic efficiency is improved. Our experiments on subtomogram classification tasks using both simulated and real data demonstrate that we can achieve comparable testing performance (on average only 3% accuracy drop) by using less than 30% of the labeled subtomograms, which shows a very promising result for subtomogram classification task with limited labeling resources.

show abstract

Section: Results and Comparisonsmentioning

confidence: 99%

Section: Active Learningmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Active Learning to Classify Macromolecular Structures in situ for Less Supervision in Cryo-Electron Tomography

Du,

Wang,

Zhu

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…However, there are still some challenges in the field. (1) Most existing methods [11,12] represent medical events as embedding vectors, which lose real value information of the medical events (e.g., lab tests and vital signs). (2) Lab tests are diagnosis-driven and therefore EHRs have lots of missing value for lab tests.…”

Section: Introductionmentioning

confidence: 99%

An interpretable risk prediction model for healthcare with pattern attention

Kamal

Yin

Qian

et al. 2020

BMC Med Inform Decis Mak

Self Cite

View full text Add to dashboard Cite

Background The availability of massive amount of data enables the possibility of clinical predictive tasks. Deep learning methods have achieved promising performance on the tasks. However, most existing methods suffer from three limitations: (1) There are lots of missing value for real value events, many methods impute the missing value and then train their models based on the imputed values, which may introduce imputation bias. The models’ performance is highly dependent on the imputation accuracy. (2) Lots of existing studies just take Boolean value medical events (e.g. diagnosis code) as inputs, but ignore real value medical events (e.g., lab tests and vital signs), which are more important for acute disease (e.g., sepsis) and mortality prediction. (3) Existing interpretable models can illustrate which medical events are conducive to the output results, but are not able to give contributions of patterns among medical events. Methods In this study, we propose a novel interpretable Pattern Attention model with Value Embedding (PAVE) to predict the risks of certain diseases. PAVE takes the embedding of various medical events, their values and the corresponding occurring time as inputs, leverage self-attention mechanism to attend to meaningful patterns among medical events for risk prediction tasks. Because only the observed values are embedded into vectors, we don’t need to impute the missing values and thus avoids the imputations bias. Moreover, the self-attention mechanism is helpful for the model interpretability, which means the proposed model can output which patterns cause high risks. Results We conduct sepsis onset prediction and mortality prediction experiments on a publicly available dataset MIMIC-III and our proprietary EHR dataset. The experimental results show that PAVE outperforms existing models. Moreover, by analyzing the self-attention weights, our model outputs meaningful medical event patterns related to mortality. Conclusions PAVE learns effective medical event representation by incorporating the values and occurring time, which can improve the risk prediction performance. Moreover, the presented self-attention mechanism can not only capture patients’ health state information, but also output the contributions of various medical event patterns, which pave the way for interpretable clinical risk predictions. Availability The code for this paper is available at: https://github.com/yinchangchang/PAVE.

show abstract

“…However, there are still some challenges in the field. (i) Most existing methods [7, 8] represent medical events as embedding vectors, which lose real value information of the medical events (e.g., lab tests and vital signs). (ii) Lab tests are diagnosis-driven and therefore EHRs have lots of missing value for lab tests.…”

Section: Introductionmentioning

confidence: 99%

An Interpretable Risk Prediction Model for Healthcare with Pattern Attention

Kamal

Yin

Qian

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Background: The availability of the massive amount of data enables the possibility of clinical predictive tasks. Deep learning methods have achieved promising performance on the tasks. However, most existing methods suffer from three limitations: (i) There are lots of missing value for real value events, many methods impute the missing value and then train their models based on the imputed values, which may introduce imputation bias. The models' performance is highly dependent on imputation accuracy. (ii) Lots of existing studies just take Boolean value medical events (e.g. diagnosis code) as inputs, but ignore real value medical events (e.g., lab tests and vital signs), which are more important for acute disease (e.g., sepsis) and mortality prediction. (iii) Existing interpretable models can illustrate which medical events are conducive to the output results, but are not able to give contributions of patterns among medical events. Methods: In this study, we propose a novel interpretable Pattern Attention model with Value Embedding (PAVE) to predict the risks of certain diseases. PAVE takes the embedding of various medical events, their values and the corresponding occurring time as inputs, leverage self-attention mechanism to attend to meaningful patterns among medical events for risk prediction tasks. Because only the observed values are embedded into vectors, we don't need to impute the missing values and thus avoids the imputations bias. Moreover, the self-attention mechanism is helpful for the model interpretability, which means the proposed model can output which patterns cause high risks. Results: We conduct sepsis onset prediction and mortality prediction experiments on a publicly available dataset MIMIC-III and our proprietary EHR dataset. The experimental results show that PAVE outperforms existing models. Moreover, by analyzing the self-attention weights, our model outputs meaningful medical event patterns related to mortality. Conclusions: PAVE learns effective medical event representation by incorporating the values and occurring time, which can improve the risk prediction performance. Moreover, the presented self-attention mechanism can not only capture patients' health state information, but also output the contributions of various medical event patterns, which pave the way for interpretable clinical risk predictions. Availability: The code for this paper is available at: https://github.com/yinchangchang/PAVE.

show abstract

Deep Similarity-Based Batch Mode Active Learning with Exploration-Exploitation

Cited by 32 publications

References 15 publications

Active Learning to Classify Macromolecular Structures in situ for Less Supervision in Cryo-Electron Tomography

Active Learning to Classify Macromolecular Structures in situ for Less Supervision in Cryo-Electron Tomography

An interpretable risk prediction model for healthcare with pattern attention

An Interpretable Risk Prediction Model for Healthcare with Pattern Attention

Contact Info

Product

Resources

About