Mengying Sun, Huijun Wang, Jing Xing, Bin Chen, Han Meng, Jiayu Zhou
Leveraging computational methods to generate small molecules with desired properties has been an active research area in the drug discovery field. Towards real-world applications, however, efficient generation of molecules that satisfy multiple property requirements simultaneously remains a key challenge. In this paper, we tackle this challenge using a search-based approach and propose a simple yet effective framework called MolSearch for multi-objective molecular generation (optimization). We show that given proper design and sufficient information, search-based methods can achieve performance comparable or even better than deep learning methods while being computationally efficient. Such efficiency enables massive exploration of chemical space given constrained computational resources. In particular, MolSearch starts with existing molecules and uses a two-stage search strategy to gradually modify them into new ones, based on transformation rules derived systematically and exhaustively from large compound libraries. We evaluate MolSearch in multiple benchmark generation settings and demonstrate its effectiveness and efficiency.
{"title":"MolSearch: Search-based Multi-objective Molecular Generation and Property Optimization.","authors":"Mengying Sun, Huijun Wang, Jing Xing, Bin Chen, Han Meng, Jiayu Zhou","doi":"10.1145/3534678.3542676","DOIUrl":"https://doi.org/10.1145/3534678.3542676","url":null,"abstract":"<p><p>Leveraging computational methods to generate small molecules with desired properties has been an active research area in the drug discovery field. Towards real-world applications, however, efficient generation of molecules that satisfy <b>multiple</b> property requirements simultaneously remains a key challenge. In this paper, we tackle this challenge using a search-based approach and propose a simple yet effective framework called MolSearch for multi-objective molecular generation (optimization). We show that given proper design and sufficient information, search-based methods can achieve performance comparable or even better than deep learning methods while being computationally efficient. Such efficiency enables massive exploration of chemical space given constrained computational resources. In particular, MolSearch starts with existing molecules and uses a two-stage search strategy to gradually modify them into new ones, based on transformation rules derived systematically and exhaustively from large compound libraries. We evaluate MolSearch in multiple benchmark generation settings and demonstrate its effectiveness and efficiency.</p>","PeriodicalId":74037,"journal":{"name":"KDD : proceedings. International Conference on Knowledge Discovery & Data Mining","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10097503/pdf/nihms-1888099.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9580527","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Despite intense efforts in basic and clinical research, an individualized ventilation strategy for critically ill patients remains a major challenge. Recently, dynamic treatment regime (DTR) with reinforcement learning (RL) on electronic health records (EHR) has attracted interest from both the healthcare industry and machine learning research community. However, most learned DTR policies might be biased due to the existence of confounders. Although some treatment actions non-survivors received may be helpful, if confounders cause the mortality, the training of RL models guided by long-term outcomes (e.g., 90-day mortality) would punish those treatment actions causing the learned DTR policies to be suboptimal. In this study, we develop a new deconfounding actor-critic network (DAC) to learn optimal DTR policies for patients. To alleviate confounding issues, we incorporate a patient resampling module and a confounding balance module into our actor-critic framework. To avoid punishing the effective treatment actions non-survivors received, we design a short-term reward to capture patients' immediate health state changes. Combining short-term with long-term rewards could further improve the model performance. Moreover, we introduce a policy adaptation method to successfully transfer the learned model to new-source small-scale datasets. The experimental results on one semi-synthetic and two different real-world datasets show the proposed model outperforms the state-of-the-art models. The proposed model provides individualized treatment decisions for mechanical ventilation that could improve patient outcomes.
{"title":"Deconfounding Actor-Critic Network with Policy Adaptation for Dynamic Treatment Regimes.","authors":"Changchang Yin, Ruoqi Liu, Jeffrey Caterino, Ping Zhang","doi":"10.1145/3534678.3539413","DOIUrl":"https://doi.org/10.1145/3534678.3539413","url":null,"abstract":"<p><p>Despite intense efforts in basic and clinical research, an individualized ventilation strategy for critically ill patients remains a major challenge. Recently, dynamic treatment regime (DTR) with reinforcement learning (RL) on electronic health records (EHR) has attracted interest from both the healthcare industry and machine learning research community. However, most learned DTR policies might be biased due to the existence of confounders. Although some treatment actions non-survivors received may be helpful, if confounders cause the mortality, the training of RL models guided by long-term outcomes (e.g., 90-day mortality) would punish those treatment actions causing the learned DTR policies to be suboptimal. In this study, we develop a new deconfounding actor-critic network (DAC) to learn optimal DTR policies for patients. To alleviate confounding issues, we incorporate a patient resampling module and a confounding balance module into our actor-critic framework. To avoid punishing the effective treatment actions non-survivors received, we design a short-term reward to capture patients' immediate health state changes. Combining short-term with long-term rewards could further improve the model performance. Moreover, we introduce a policy adaptation method to successfully transfer the learned model to new-source small-scale datasets. The experimental results on one semi-synthetic and two different real-world datasets show the proposed model outperforms the state-of-the-art models. The proposed model provides individualized treatment decisions for mechanical ventilation that could improve patient outcomes.</p>","PeriodicalId":74037,"journal":{"name":"KDD : proceedings. International Conference on Knowledge Discovery & Data Mining","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9466407/pdf/nihms-1830314.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"40354004","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Age-related macular degeneration (AMD) is the leading cause of irreversible blindness in developed countries. Identifying patients at high risk of progression to late AMD, the sight-threatening stage, is critical for clinical actions, including medical interventions and timely monitoring. Recently, deep-learning-based models have been developed and achieved superior performance for late AMD prediction. However, most existing methods are limited to the color fundus photography (CFP) from the last ophthalmic visit and do not include the longitudinal CFP history and AMD progression during the previous years' visits. Patients in different AMD subphenotypes might have various speeds of progression in different stages of AMD disease. Capturing the progression information during the previous years' visits might be useful for the prediction of AMD progression. In this work, we propose a Contrastive-Attention-based Time-aware Long Short-Term Memory network (CAT-LSTM) to predict AMD progression. First, we adopt a convolutional neural network (CNN) model with a contrastive attention module (CA) to extract abnormal features from CFPs. Then we utilize a time-aware LSTM (T-LSTM) to model the patients' history and consider the AMD progression information. The combination of disease progression, genotype information, demographics, and CFP features are sent to T-LSTM. Moreover, we leverage an auto-encoder to represent temporal CFP sequences as fixed-size vectors and adopt k-means to cluster them into subphenotypes. We evaluate the proposed model based on real-world datasets, and the results show that the proposed model could achieve 0.925 on area under the receiver operating characteristic (AUROC) for 5-year late-AMD prediction and outperforms the state-of-the-art methods by more than 3%, which demonstrates the effectiveness of the proposed CAT-LSTM. After analyzing patient representation learned by an auto-encoder, we identify 3 novel subphenotypes of AMD patients with different characteristics and progression rates to late AMD, paving the way for improved personalization of AMD management. The code of CAT-LSTM can be found at GitHub.
{"title":"Predicting Age-Related Macular Degeneration Progression with Contrastive Attention and Time-Aware LSTM.","authors":"Changchang Yin, Sayoko E Moroi, Ping Zhang","doi":"10.1145/3534678.3539163","DOIUrl":"https://doi.org/10.1145/3534678.3539163","url":null,"abstract":"<p><p>Age-related macular degeneration (AMD) is the leading cause of irreversible blindness in developed countries. Identifying patients at high risk of progression to late AMD, the sight-threatening stage, is critical for clinical actions, including medical interventions and timely monitoring. Recently, deep-learning-based models have been developed and achieved superior performance for late AMD prediction. However, most existing methods are limited to the color fundus photography (CFP) from the last ophthalmic visit and do not include the longitudinal CFP history and AMD progression during the previous years' visits. Patients in different AMD subphenotypes might have various speeds of progression in different stages of AMD disease. Capturing the progression information during the previous years' visits might be useful for the prediction of AMD progression. In this work, we propose a <b>C</b>ontrastive-<b>A</b>ttention-based <b>T</b>ime-aware <b>L</b>ong <b>S</b>hort-<b>T</b>erm <b>M</b>emory network (<b>CAT-LSTM</b>) to predict AMD progression. First, we adopt a convolutional neural network (CNN) model with a contrastive attention module (CA) to extract abnormal features from CFPs. Then we utilize a time-aware LSTM (T-LSTM) to model the patients' history and consider the AMD progression information. The combination of disease progression, genotype information, demographics, and CFP features are sent to T-LSTM. Moreover, we leverage an auto-encoder to represent temporal CFP sequences as fixed-size vectors and adopt k-means to cluster them into subphenotypes. We evaluate the proposed model based on real-world datasets, and the results show that the proposed model could achieve 0.925 on area under the receiver operating characteristic (AUROC) for 5-year late-AMD prediction and outperforms the state-of-the-art methods by more than 3%, which demonstrates the effectiveness of the proposed CAT-LSTM. After analyzing patient representation learned by an auto-encoder, we identify 3 novel subphenotypes of AMD patients with different characteristics and progression rates to late AMD, paving the way for improved personalization of AMD management. The code of CAT-LSTM can be found at GitHub.</p>","PeriodicalId":74037,"journal":{"name":"KDD : proceedings. International Conference on Knowledge Discovery & Data Mining","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9505703/pdf/nihms-1830315.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9234125","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-08-01Epub Date: 2021-08-14DOI: 10.1145/3447548.3467186
Mengying Sun, Jing Xing, Huijun Wang, Bin Chen, Jiayu Zhou
Recent years have seen a rapid growth of utilizing graph neural networks (GNNs) in the biomedical domain for tackling drug-related problems. However, like any other deep architectures, GNNs are data hungry. While requiring labels in real world is often expensive, pretraining GNNs in an unsupervised manner has been actively explored. Among them, graph contrastive learning, by maximizing the mutual information between paired graph augmentations, has been shown to be effective on various downstream tasks. However, the current graph contrastive learning framework has two limitations. First, the augmentations are designed for general graphs and thus may not be suitable or powerful enough for certain domains. Second, the contrastive scheme only learns representations that are invariant to local perturbations and thus does not consider the global structure of the dataset, which may also be useful for downstream tasks. In this paper, we study graph contrastive learning designed specifically for the biomedical domain, where molecular graphs are present. We propose a novel framework called MoCL, which utilizes domain knowledge at both local- and global-level to assist representation learning. The local-level domain knowledge guides the augmentation process such that variation is introduced without changing graph semantics. The global-level knowledge encodes the similarity information between graphs in the entire dataset and helps to learn representations with richer semantics. The entire model is learned through a double contrast objective. We evaluate MoCL on various molecular datasets under both linear and semi-supervised settings and results show that MoCL achieves state-of-the-art performance.
{"title":"MoCL: Data-driven Molecular Fingerprint via Knowledge-aware Contrastive Learning from Molecular Graph.","authors":"Mengying Sun, Jing Xing, Huijun Wang, Bin Chen, Jiayu Zhou","doi":"10.1145/3447548.3467186","DOIUrl":"10.1145/3447548.3467186","url":null,"abstract":"<p><p>Recent years have seen a rapid growth of utilizing graph neural networks (GNNs) in the biomedical domain for tackling drug-related problems. However, like any other deep architectures, GNNs are data hungry. While requiring labels in real world is often expensive, pretraining GNNs in an unsupervised manner has been actively explored. Among them, graph contrastive learning, by maximizing the mutual information between paired graph augmentations, has been shown to be effective on various downstream tasks. However, the current graph contrastive learning framework has two limitations. First, the augmentations are designed for general graphs and thus may not be suitable or powerful enough for certain domains. Second, the contrastive scheme only learns representations that are invariant to local perturbations and thus does not consider the global structure of the dataset, which may also be useful for downstream tasks. In this paper, we study graph contrastive learning designed specifically for the biomedical domain, where molecular graphs are present. We propose a novel framework called MoCL, which utilizes domain knowledge at both local- and global-level to assist representation learning. The local-level domain knowledge guides the augmentation process such that variation is introduced without changing graph semantics. The global-level knowledge encodes the similarity information between graphs in the entire dataset and helps to learn representations with richer semantics. The entire model is learned through a double contrast objective. We evaluate MoCL on various molecular datasets under both linear and semi-supervised settings and results show that MoCL achieves state-of-the-art performance.</p>","PeriodicalId":74037,"journal":{"name":"KDD : proceedings. International Conference on Knowledge Discovery & Data Mining","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9105980/pdf/nihms-1798075.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"10249436","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Federated learning is a distributed learning framework that is communication efficient and provides protection over participating users' raw training data. One outstanding challenge of federate learning comes from the users' heterogeneity, and learning from such data may yield biased and unfair models for minority groups. While adversarial learning is commonly used in centralized learning for mitigating bias, there are significant barriers when extending it to the federated framework. In this work, we study these barriers and address them by proposing a novel approach Federated Adversarial DEbiasing (FADE). FADE does not require users' sensitive group information for debiasing and offers users the freedom to opt-out from the adversarial component when privacy or computational costs become a concern. We show that ideally, FADE can attain the same global optimality as the one by the centralized algorithm. We then analyze when its convergence may fail in practice and propose a simple yet effective method to address the problem. Finally, we demonstrate the effectiveness of the proposed framework through extensive empirical studies, including the problem settings of unsupervised domain adaptation and fair learning. Our codes and pre-trained models are available at: https://github.com/illidanlab/FADE.
{"title":"Federated Adversarial Debiasing for Fair and Transferable Representations.","authors":"Junyuan Hong, Zhuangdi Zhu, Shuyang Yu, Zhangyang Wang, Hiroko Dodge, Jiayu Zhou","doi":"10.1145/3447548.3467281","DOIUrl":"10.1145/3447548.3467281","url":null,"abstract":"<p><p>Federated learning is a distributed learning framework that is communication efficient and provides protection over participating users' raw training data. One outstanding challenge of federate learning comes from the users' heterogeneity, and learning from such data may yield biased and unfair models for minority groups. While adversarial learning is commonly used in centralized learning for mitigating bias, there are significant barriers when extending it to the federated framework. In this work, we study these barriers and address them by proposing a novel approach Federated Adversarial DEbiasing (FADE). FADE does not require users' sensitive group information for debiasing and offers users the freedom to opt-out from the adversarial component when privacy or computational costs become a concern. We show that ideally, FADE can attain the same global optimality as the one by the centralized algorithm. We then analyze when its convergence may fail in practice and propose a simple yet effective method to address the problem. Finally, we demonstrate the effectiveness of the proposed framework through extensive empirical studies, including the problem settings of unsupervised domain adaptation and fair learning. Our codes and pre-trained models are available at: https://github.com/illidanlab/FADE.</p>","PeriodicalId":74037,"journal":{"name":"KDD : proceedings. International Conference on Knowledge Discovery & Data Mining","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9105979/pdf/nihms-1798074.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"10249439","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Kejing Yin, Ardavan Afshar, Joyce C Ho, William K Cheung, Chao Zhang, Jimeng Sun
Binary data with one-class missing values are ubiquitous in real-world applications. They can be represented by irregular tensors with varying sizes in one dimension, where value one means presence of a feature while zero means unknown (i.e., either presence or absence of a feature). Learning accurate low-rank approximations from such binary irregular tensors is a challenging task. However, none of the existing models developed for factorizing irregular tensors take the missing values into account, and they assume Gaussian distributions, resulting in a distribution mismatch when applied to binary data. In this paper, we propose Logistic PARAFAC2 (LogPar) by modeling the binary irregular tensor with Bernoulli distribution parameterized by an underlying real-valued tensor. Then we approximate the underlying tensor with a positive-unlabeled learning loss function to account for the missing values. We also incorporate uniqueness and temporal smoothness regularization to enhance the interpretability. Extensive experiments using large-scale real-world datasets show that LogPar outperforms all baselines in both irregular tensor completion and downstream predictive tasks. For the irregular tensor completion, LogPar achieves up to 26% relative improvement compared to the best baseline. Besides, LogPar obtains relative improvement of 13.2% for heart failure prediction and 14% for mortality prediction on average compared to the state-of-the-art PARAFAC2 models.
{"title":"LogPar: Logistic PARAFAC2 Factorization for Temporal Binary Data with Missing Values.","authors":"Kejing Yin, Ardavan Afshar, Joyce C Ho, William K Cheung, Chao Zhang, Jimeng Sun","doi":"10.1145/3394486.3403213","DOIUrl":"https://doi.org/10.1145/3394486.3403213","url":null,"abstract":"<p><p>Binary data with one-class missing values are ubiquitous in real-world applications. They can be represented by irregular tensors with varying sizes in one dimension, where value one means presence of a feature while zero means unknown (i.e., either presence or absence of a feature). Learning accurate low-rank approximations from such binary irregular tensors is a challenging task. However, none of the existing models developed for factorizing irregular tensors take the missing values into account, and they assume Gaussian distributions, resulting in a distribution mismatch when applied to binary data. In this paper, we propose Logistic PARAFAC2 (LogPar) by modeling the binary irregular tensor with Bernoulli distribution parameterized by an underlying real-valued tensor. Then we approximate the underlying tensor with a positive-unlabeled learning loss function to account for the missing values. We also incorporate uniqueness and temporal smoothness regularization to enhance the interpretability. Extensive experiments using large-scale real-world datasets show that LogPar outperforms all baselines in both irregular tensor completion and downstream predictive tasks. For the irregular tensor completion, LogPar achieves up to 26% relative improvement compared to the best baseline. Besides, LogPar obtains relative improvement of 13.2% for heart failure prediction and 14% for mortality prediction on average compared to the state-of-the-art PARAFAC2 models.</p>","PeriodicalId":74037,"journal":{"name":"KDD : proceedings. International Conference on Knowledge Discovery & Data Mining","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1145/3394486.3403213","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"39079152","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Xi Sheryl Zhang, Fengyi Tang, Hiroko H Dodge, Jiayu Zhou, Fei Wang
In recent years, large amounts of health data, such as patient Electronic Health Records (EHR), are becoming readily available. This provides an unprecedented opportunity for knowledge discovery and data mining algorithms to dig insights from them, which can, later on, be helpful to the improvement of the quality of care delivery. Predictive modeling of clinical risks, including in-hospital mortality, hospital readmission, chronic disease onset, condition exacerbation, etc., from patient EHR, is one of the health data analytic problems that attract lots of the interests. The reason is not only because the problem is important in clinical settings, but also is challenging when working with EHR such as sparsity, irregularity, temporality, etc. Different from applications in other domains such as computer vision and natural language processing, the data samples in medicine (patients) are relatively limited, which creates lots of troubles for building effective predictive models, especially for complicated ones such as deep learning. In this paper, we propose MetaPred, a meta-learning framework for clinical risk prediction from longitudinal patient EHR. In particular, in order to predict the target risk with limited data samples, we train a meta-learner from a set of related risk prediction tasks which learns how a good predictor is trained. The meta-learned can then be directly used in target risk prediction, and the limited available samples in the target domain can be used for further fine-tuning the model performance. The effectiveness of MetaPred is tested on a real patient EHR repository from Oregon Health & Science University. We are able to demonstrate that with Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN) as base predictors, MetaPred can achieve much better performance for predicting target risk with low resources comparing with the predictor trained on the limited samples available for this risk alone.
{"title":"MetaPred: Meta-Learning for Clinical Risk Prediction with Limited Patient Electronic Health Records.","authors":"Xi Sheryl Zhang, Fengyi Tang, Hiroko H Dodge, Jiayu Zhou, Fei Wang","doi":"10.1145/3292500.3330779","DOIUrl":"https://doi.org/10.1145/3292500.3330779","url":null,"abstract":"<p><p>In recent years, large amounts of health data, such as patient Electronic Health Records (EHR), are becoming readily available. This provides an unprecedented opportunity for knowledge discovery and data mining algorithms to dig insights from them, which can, later on, be helpful to the improvement of the quality of care delivery. Predictive modeling of clinical risks, including in-hospital mortality, hospital readmission, chronic disease onset, condition exacerbation, etc., from patient EHR, is one of the health data analytic problems that attract lots of the interests. The reason is not only because the problem is important in clinical settings, but also is challenging when working with EHR such as sparsity, irregularity, temporality, etc. Different from applications in other domains such as computer vision and natural language processing, the data samples in medicine (patients) are relatively limited, which creates lots of troubles for building effective predictive models, especially for complicated ones such as deep learning. In this paper, we propose MetaPred, a meta-learning framework for clinical risk prediction from longitudinal patient EHR. In particular, in order to predict the target risk with limited data samples, we train a meta-learner from a set of related risk prediction tasks which learns how a good predictor is trained. The meta-learned can then be directly used in target risk prediction, and the limited available samples in the target domain can be used for further fine-tuning the model performance. The effectiveness of MetaPred is tested on a real patient EHR repository from Oregon Health & Science University. We are able to demonstrate that with Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN) as base predictors, MetaPred can achieve much better performance for predicting target risk with low resources comparing with the predictor trained on the limited samples available for this risk alone.</p>","PeriodicalId":74037,"journal":{"name":"KDD : proceedings. International Conference on Knowledge Discovery & Data Mining","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1145/3292500.3330779","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"38879115","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Sho Inaba, Carl T Fakhry, Rahul V Kulkarni, Kourosh Zarringhalam
We present a reformulation of the distance metric learning problem as a penalized optimization problem, with a penalty term corresponding to the von Neumann entropy of the distance metric. This formulation leads to a mapping to statistical mechanics such that the metric learning optimization problem becomes equivalent to free energy minimization. Correspondingly, our approach leads to an analytical solution of the optimization problem based on the Boltzmann distribution. The mapping established in this work suggests new approaches for dimensionality reduction and provides insights into determination of optimal parameters for the penalty term. Furthermore, we demonstrate that the metric projects the data onto direction of maximum dissimilarity with optimal and tunable separation between classes and thus the transformation can be used for high dimensional data visualization, classification, and clustering tasks. We benchmark our method against previous distance learning methods and provide an efficient implementation in an R package available to download at: https://github.com/kouroshz/fenn.
{"title":"A Free Energy Based Approach for Distance Metric Learning.","authors":"Sho Inaba, Carl T Fakhry, Rahul V Kulkarni, Kourosh Zarringhalam","doi":"10.1145/3292500.3330975","DOIUrl":"https://doi.org/10.1145/3292500.3330975","url":null,"abstract":"<p><p>We present a reformulation of the distance metric learning problem as a penalized optimization problem, with a penalty term corresponding to the von Neumann entropy of the distance metric. This formulation leads to a mapping to statistical mechanics such that the metric learning optimization problem becomes equivalent to free energy minimization. Correspondingly, our approach leads to an analytical solution of the optimization problem based on the Boltzmann distribution. The mapping established in this work suggests new approaches for dimensionality reduction and provides insights into determination of optimal parameters for the penalty term. Furthermore, we demonstrate that the metric projects the data onto direction of maximum dissimilarity with optimal and tunable separation between classes and thus the transformation can be used for high dimensional data visualization, classification, and clustering tasks. We benchmark our method against previous distance learning methods and provide an efficient implementation in an R package available to download at: https://github.com/kouroshz/fenn.</p>","PeriodicalId":74037,"journal":{"name":"KDD : proceedings. International Conference on Knowledge Discovery & Data Mining","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1145/3292500.3330975","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"25467271","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Fengyi Tang, Cao Xiao, Fei Wang, Jiayu Zhou, Li-Wei H Lehman
Knowledge transfer has been of great interest in current machine learning research, as many have speculated its importance in modeling the human ability to rapidly generalize learned models to new scenarios. Particularly in cases where training samples are limited, knowledge transfer shows improvement on both the learning speed and generalization performance of related tasks. Recently, Learning Using Privileged Information (LUPI) has presented a new direction in knowledge transfer by modeling the transfer of prior knowledge as a Teacher-Student interaction process. Under LUPI, a Teacher model uses Privileged Information (PI) that is only available at training time to improve the sample complexity required to train a Student learner for a given task. In this work, we present a LUPI formulation that allows privileged information to be retained in a multi-task learning setting. We propose a novel feature matching algorithm that projects samples from the original feature space and the privilege information space into a joint latent space in a way that informs similarity between training samples. Our experiments show that useful knowledge from PI is maintained in the latent space and greatly improves the sample efficiency of other related learning tasks. We also provide an analysis of sample complexity of the proposed LUPI method, which under some favorable assumptions can achieve a greater sample efficiency than brute force methods.
{"title":"Retaining Privileged Information for Multi-Task Learning.","authors":"Fengyi Tang, Cao Xiao, Fei Wang, Jiayu Zhou, Li-Wei H Lehman","doi":"10.1145/3292500.3330907","DOIUrl":"https://doi.org/10.1145/3292500.3330907","url":null,"abstract":"<p><p>Knowledge transfer has been of great interest in current machine learning research, as many have speculated its importance in modeling the human ability to rapidly generalize learned models to new scenarios. Particularly in cases where training samples are limited, knowledge transfer shows improvement on both the learning speed and generalization performance of related tasks. Recently, <i>Learning Using Privileged Information</i> (LUPI) has presented a new direction in knowledge transfer by modeling the transfer of prior knowledge as a Teacher-Student interaction process. Under LUPI, a Teacher model uses Privileged Information (PI) that is only available at training time to improve the sample complexity required to train a Student learner for a given task. In this work, we present a LUPI formulation that allows privileged information to be retained in a multi-task learning setting. We propose a novel feature matching algorithm that projects samples from the original feature space and the privilege information space into a <i>joint latent space</i> in a way that informs similarity between training samples. Our experiments show that useful knowledge from PI is maintained in the latent space and greatly improves the sample efficiency of other related learning tasks. We also provide an analysis of sample complexity of the proposed LUPI method, which under some favorable assumptions can achieve a greater sample efficiency than brute force methods.</p>","PeriodicalId":74037,"journal":{"name":"KDD : proceedings. International Conference on Knowledge Discovery & Data Mining","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1145/3292500.3330907","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"39890793","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yongjun Chen, Min Shi, Hongyang Gao, Dinggang Shen, Lei Cai, Shuiwang Ji
Deep learning methods have shown great success in pixel-wise prediction tasks. One of the most popular methods employs an encoder-decoder network in which deconvolutional layers are used for up-sampling feature maps. However, a key limitation of the deconvolutional layer is that it suffers from the checkerboard artifact problem, which harms the prediction accuracy. This is caused by the independency among adjacent pixels on the output feature maps. Previous work only solved the checkerboard artifact issue of deconvolutional layers in the 2D space. Since the number of intermediate feature maps needed to generate a deconvolutional layer grows exponentially with dimensionality, it is more challenging to solve this issue in higher dimensions. In this work, we propose the voxel deconvolutional layer (VoxelDCL) to solve the checkerboard artifact problem of deconvolutional layers in 3D space. We also provide an efficient approach to implement VoxelDCL. To demonstrate the effectiveness of VoxelDCL, we build four variations of voxel deconvolutional networks (VoxelDCN) based on the U-Net architecture with VoxelDCL. We apply our networks to address volumetric brain images labeling tasks using the ADNI and LONI LPBA40 datasets. The experimental results show that the proposed iVoxelDCNa achieves improved performance in all experiments. It reaches 83.34% in terms of dice ratio on the ADNI dataset and 79.12% on the LONI LPBA40 dataset, which increases 1.39% and 2.21% respectively compared with the baseline. In addition, all the variations of VoxelDCN we proposed outperform the baseline methods on the above datasets, which demonstrates the effectiveness of our methods.
{"title":"Voxel Deconvolutional Networks for 3D Brain Image Labeling.","authors":"Yongjun Chen, Min Shi, Hongyang Gao, Dinggang Shen, Lei Cai, Shuiwang Ji","doi":"10.1145/3219819.3219974","DOIUrl":"10.1145/3219819.3219974","url":null,"abstract":"<p><p>Deep learning methods have shown great success in pixel-wise prediction tasks. One of the most popular methods employs an encoder-decoder network in which deconvolutional layers are used for up-sampling feature maps. However, a key limitation of the deconvolutional layer is that it suffers from the checkerboard artifact problem, which harms the prediction accuracy. This is caused by the independency among adjacent pixels on the output feature maps. Previous work only solved the checkerboard artifact issue of deconvolutional layers in the 2D space. Since the number of intermediate feature maps needed to generate a deconvolutional layer grows exponentially with dimensionality, it is more challenging to solve this issue in higher dimensions. In this work, we propose the voxel deconvolutional layer (VoxelDCL) to solve the checkerboard artifact problem of deconvolutional layers in 3D space. We also provide an efficient approach to implement VoxelDCL. To demonstrate the effectiveness of VoxelDCL, we build four variations of voxel deconvolutional networks (VoxelDCN) based on the U-Net architecture with VoxelDCL. We apply our networks to address volumetric brain images labeling tasks using the ADNI and LONI LPBA40 datasets. The experimental results show that the proposed iVoxelDCNa achieves improved performance in all experiments. It reaches 83.34% in terms of dice ratio on the ADNI dataset and 79.12% on the LONI LPBA40 dataset, which increases 1.39% and 2.21% respectively compared with the baseline. In addition, all the variations of VoxelDCN we proposed outperform the baseline methods on the above datasets, which demonstrates the effectiveness of our methods.</p>","PeriodicalId":74037,"journal":{"name":"KDD : proceedings. International Conference on Knowledge Discovery & Data Mining","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6426146/pdf/nihms-988372.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"37087638","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}