{"title":"Unsupervised Stain Decomposition via Inversion Regulation for Multiplex Immunohistochemistry Images.","authors":"Shahira Abousamra, Danielle Fassler, Jiachen Yao, Rajarsi Gupta, Tahsin Kurc, Luisa Escobar-Hoyos, Dimitris Samaras, Kenneth Shroyer, Joel Saltz, Chao Chen","doi":"","DOIUrl":"","url":null,"abstract":"<p><p>Multiplex Immunohistochemistry (mIHC) is a cost-effective and accessible method for in situ labeling of multiple protein biomarkers in a tissue sample. By assigning a different stain to each biomarker, it allows the visualization of different types of cells within the tumor vicinity for downstream analysis. However, to detect different types of stains in a given mIHC image is a challenging problem, especially when the number of stains is high. Previous deep-learning-based methods mostly assume full supervision; yet the annotation can be costly. In this paper, we propose a novel unsupervised stain decomposition method to detect different stains simultaneously. Our method does not require any supervision, except for color samples of different stains. A main technical challenge is that the problem is underdetermined and can have multiple solutions. To conquer this issue, we propose a novel inversion regulation technique, which eliminates most undesirable solutions. On a 7-plexed IHC images dataset, the proposed method achieves high quality stain decomposition results without human annotation.</p>","PeriodicalId":74504,"journal":{"name":"Proceedings of machine learning research","volume":"227 ","pages":"74-94"},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11138139/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141181231","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}
{"title":"On Distribution Dependent Sub-Logarithmic Query Time of Learned Indexing.","authors":"Sepanta Zeighami, Cyrus Shahabi","doi":"","DOIUrl":"","url":null,"abstract":"<p><p>A fundamental problem in data management is to find the elements in an array that match a query. Recently, learned indexes are being extensively used to solve this problem, where they learn a model to predict the location of the items in the array. They are empirically shown to outperform non-learned methods (e.g., B-trees or binary search that answer queries in <math><mi>O</mi><mo>(</mo><mi>l</mi><mi>o</mi><mi>g</mi><mspace></mspace><mi>n</mi><mo>)</mo></math> time) by orders of magnitude. However, success of learned indexes has not been theoretically justified. Only existing attempt shows the same query time of <math><mi>O</mi><mo>(</mo><mi>l</mi><mi>o</mi><mi>g</mi><mspace></mspace><mi>n</mi><mo>)</mo></math>, but with a constant factor improvement in space complexity over non-learned methods, under some assumptions on data distribution. In this paper, we significantly strengthen this result, showing that under mild assumptions on data distribution, and the same space complexity as non-learned methods, learned indexes can answer queries in <math><mi>O</mi><mo>(</mo><mi>l</mi><mi>o</mi><mi>g</mi><mi>l</mi><mi>o</mi><mi>g</mi><mspace></mspace><mi>n</mi><mo>)</mo></math> expected query time. We also show that allowing for slightly larger but still near-linear space overhead, a learned index can achieve <math><mi>O</mi><mo>(</mo><mn>1</mn><mo>)</mo></math> expected query time. Our results theoretically prove learned indexes are orders of magnitude faster than non-learned methods, theoretically grounding their empirical success.</p>","PeriodicalId":74504,"journal":{"name":"Proceedings of machine learning research","volume":"202 ","pages":"40669-40680"},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10627073/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"71489774","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}
{"title":"Estimating Causal Effects using a Multi-task Deep Ensemble.","authors":"Ziyang Jiang, Zhuoran Hou, Yiling Liu, Yiman Ren, Keyu Li, David Carlson","doi":"","DOIUrl":"","url":null,"abstract":"<p><p>A number of methods have been proposed for causal effect estimation, yet few have demonstrated efficacy in handling data with complex structures, such as images. To fill this gap, we propose Causal Multi-task Deep Ensemble (CMDE), a novel framework that learns both shared and group-specific information from the study population. We provide proofs demonstrating equivalency of CDME to a multi-task Gaussian process (GP) with a coregionalization kernel <i>a priori</i>. Compared to multi-task GP, CMDE efficiently handles high-dimensional and multi-modal covariates and provides pointwise uncertainty estimates of causal effects. We evaluate our method across various types of datasets and tasks and find that CMDE outperforms state-of-the-art methods on a majority of these tasks.</p>","PeriodicalId":74504,"journal":{"name":"Proceedings of machine learning research","volume":"202 ","pages":"15023-15040"},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10759931/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139089657","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}
{"title":"Actor-Critic Alignment for Offline-to-Online Reinforcement Learning.","authors":"Zishun Yu, Xinhua Zhang","doi":"","DOIUrl":"","url":null,"abstract":"<p><p>Deep offline reinforcement learning has recently demonstrated considerable promises in leveraging offline datasets, providing high-quality models that significantly reduce the online interactions required for fine-tuning. However, such a benefit is often diminished due to the marked state-action distribution shift, which causes significant bootstrap error and wipes out the good initial policy Existing solutions resort to constraining the policy shift or balancing the sample replay based on their online-ness. However, they require online estimation of distribution divergence or density ratio. To avoid such complications, we propose deviating from existing actor-critic approaches that directly transfer the state-action value functions. Instead, we post-process them by aligning with the offline learned policy, so that the <math><mi>Q</mi></math> -values for actions outside the offline policy are also tamed. As a result, the online fine-tuning can be simply performed as in the standard actor-critic algorithms. We show empirically that the proposed method improves the performance of the fine-tuned robotic agents on various simulated tasks.</p>","PeriodicalId":74504,"journal":{"name":"Proceedings of machine learning research","volume":"202 ","pages":"40452-40474"},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11232493/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141565256","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}
Sarah Rathnam, S. Parbhoo, Weiwei Pan, Susan A. Murphy, F. Doshi-Velez
{"title":"The Unintended Consequences of Discount Regularization: Improving Regularization in Certainty Equivalence Reinforcement Learning","authors":"Sarah Rathnam, S. Parbhoo, Weiwei Pan, Susan A. Murphy, F. Doshi-Velez","doi":"10.48550/arXiv.2306.11208","DOIUrl":"https://doi.org/10.48550/arXiv.2306.11208","url":null,"abstract":"Discount regularization, using a shorter planning horizon when calculating the optimal policy, is a popular choice to restrict planning to a less complex set of policies when estimating an MDP from sparse or noisy data (Jiang et al., 2015). It is commonly understood that discount regularization functions by de-emphasizing or ignoring delayed effects. In this paper, we reveal an alternate view of discount regularization that exposes unintended consequences. We demonstrate that planning under a lower discount factor produces an identical optimal policy to planning using any prior on the transition matrix that has the same distribution for all states and actions. In fact, it functions like a prior with stronger regularization on state-action pairs with more transition data. This leads to poor performance when the transition matrix is estimated from data sets with uneven amounts of data across state-action pairs. Our equivalence theorem leads to an explicit formula to set regularization parameters locally for individual state-action pairs rather than globally. We demonstrate the failures of discount regularization and how we remedy them using our state-action-specific method across simple empirical examples as well as a medical cancer simulator.","PeriodicalId":74504,"journal":{"name":"Proceedings of machine learning research","volume":"202 1","pages":"28746-28767"},"PeriodicalIF":0.0,"publicationDate":"2023-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47482371","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"On Distribution Dependent Sub-Logarithmic Query Time of Learned Indexing","authors":"Sepanta Zeighami, C. Shahabi","doi":"10.48550/arXiv.2306.10651","DOIUrl":"https://doi.org/10.48550/arXiv.2306.10651","url":null,"abstract":"A fundamental problem in data management is to find the elements in an array that match a query. Recently, learned indexes are being extensively used to solve this problem, where they learn a model to predict the location of the items in the array. They are empirically shown to outperform non-learned methods (e.g., B-trees or binary search that answer queries in O(logn) time) by orders of magnitude. However, success of learned indexes has not been theoretically justified. Only existing attempt shows the same query time of O(logn), but with a constant factor improvement in space complexity over non-learned methods, under some assumptions on data distribution. In this paper, we significantly strengthen this result, showing that under mild assumptions on data distribution, and the same space complexity as non-learned methods, learned indexes can answer queries in O(loglogn) expected query time. We also show that allowing for slightly larger but still near-linear space overhead, a learned index can achieve O(1) expected query time. Our results theoretically prove learned indexes are orders of magnitude faster than non-learned methods, theoretically grounding their empirical success.","PeriodicalId":74504,"journal":{"name":"Proceedings of machine learning research","volume":"261 1","pages":"40669-40680"},"PeriodicalIF":0.0,"publicationDate":"2023-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79545250","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Karine Karine, P. Klasnja, Susan A. Murphy, Benjamin M Marlin
{"title":"Assessing the Impact of Context Inference Error and Partial Observability on RL Methods for Just-In-Time Adaptive Interventions","authors":"Karine Karine, P. Klasnja, Susan A. Murphy, Benjamin M Marlin","doi":"10.48550/arXiv.2305.09913","DOIUrl":"https://doi.org/10.48550/arXiv.2305.09913","url":null,"abstract":"Just-in-Time Adaptive Interventions (JITAIs) are a class of personalized health interventions developed within the behavioral science community. JITAIs aim to provide the right type and amount of support by iteratively selecting a sequence of intervention options from a pre-defined set of components in response to each individual's time varying state. In this work, we explore the application of reinforcement learning methods to the problem of learning intervention option selection policies. We study the effect of context inference error and partial observability on the ability to learn effective policies. Our results show that the propagation of uncertainty from context inferences is critical to improving intervention efficacy as context uncertainty increases, while policy gradient algorithms can provide remarkable robustness to partially observed behavioral state information.","PeriodicalId":74504,"journal":{"name":"Proceedings of machine learning research","volume":"216 1","pages":"1047-1057"},"PeriodicalIF":0.0,"publicationDate":"2023-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46832340","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"DIET: Conditional independence testing with marginal dependence measures of residual information.","authors":"Mukund Sudarshan, Aahlad Puli, Wesley Tansey, Rajesh Ranganath","doi":"","DOIUrl":"","url":null,"abstract":"<p><p>Conditional randomization tests (CRTs) assess whether a variable <math><mi>x</mi></math> is predictive of another variable <math><mi>y</mi></math>, having observed covariates <math><mrow><mi>z</mi></mrow></math>. CRTs require fitting a large number of predictive models, which is often computationally intractable. Existing solutions to reduce the cost of CRTs typically split the dataset into a train and test portion, or rely on heuristics for interactions, both of which lead to a loss in power. We propose the decoupled independence test (DIET), an algorithm that avoids both of these issues by leveraging marginal independence statistics to test conditional independence relationships. DIET tests the marginal independence of two random variables: <math><mrow><msub><mi>F</mi><mrow><mi>x</mi><mo>∣</mo><mi>z</mi></mrow></msub><mo>(</mo><mi>x</mi><mo>∣</mo><mi>z</mi><mo>)</mo></mrow></math> and <math><mrow><msub><mi>F</mi><mrow><mi>y</mi><mo>∣</mo><mi>z</mi></mrow></msub><mo>(</mo><mi>y</mi><mo>∣</mo><mi>z</mi><mo>)</mo></mrow></math> where <math><mrow><msub><mi>F</mi><mrow><mo>⋅</mo><mo>∣</mo><mi>z</mi></mrow></msub><mo>(</mo><mo>⋅</mo><mo>∣</mo><mi>z</mi><mo>)</mo></mrow></math> is a conditional cumulative distribution function (CDF) for the distribution <math><mrow><mi>p</mi><mo>(</mo><mo>⋅</mo><mo>∣</mo><mi>z</mi><mo>)</mo></mrow></math>. These variables are termed \"information residuals.\" We give sufficient conditions for DIET to achieve finite sample type-1 error control and power greater than the type-1 error rate. We then prove that when using the mutual information between the information residuals as a test statistic, DIET yields the most powerful conditionally valid test. Finally, we show DIET achieves higher power than other tractable CRTs on several synthetic and real benchmarks.</p>","PeriodicalId":74504,"journal":{"name":"Proceedings of machine learning research","volume":"206 ","pages":"10343-10367"},"PeriodicalIF":0.0,"publicationDate":"2023-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10484293/pdf/nihms-1899844.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"10577745","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}
{"title":"Covariate-informed Representation Learning to Prevent Posterior Collapse of iVAE.","authors":"Young-Geun Kim, Ying Liu, Xue-Xin Wei","doi":"","DOIUrl":"","url":null,"abstract":"<p><p>The recently proposed identifiable variational autoencoder (iVAE) framework provides a promising approach for learning latent independent components (ICs). iVAEs use auxiliary covariates to build an identifiable generation structure from covariates to ICs to observations, and the posterior network approximates ICs given observations and covariates. Though the identifiability is appealing, we show that iVAEs could have local minimum solution where observations and the approximated ICs are independent given covariates.-a phenomenon we referred to as the posterior collapse problem of iVAEs. To overcome this problem, we develop a new approach, covariate-informed iVAE (CI-iVAE) by considering a mixture of encoder and posterior distributions in the objective function. In doing so, the objective function prevents the posterior collapse, resulting latent representations that contain more information of the observations. Furthermore, CI-iVAE extends the original iVAE objective function to a larger class and finds the optimal one among them, thus having tighter evidence lower bounds than the original iVAE. Experiments on simulation datasets, EMNIST, Fashion-MNIST, and a large-scale brain imaging dataset demonstrate the effectiveness of our new method.</p>","PeriodicalId":74504,"journal":{"name":"Proceedings of machine learning research","volume":"206 ","pages":"2641-2660"},"PeriodicalIF":0.0,"publicationDate":"2023-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10226378/pdf/nihms-1902106.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9908011","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}
{"title":"Don't be fooled: label leakage in explanation methods and the importance of their quantitative evaluation.","authors":"Neil Jethani, Adriel Saporta, Rajesh Ranganath","doi":"","DOIUrl":"","url":null,"abstract":"<p><p>Feature attribution methods identify which features of an input most influence a model's output. Most widely-used feature attribution methods (such as SHAP, LIME, and Grad-CAM) are \"class-dependent\" methods in that they generate a feature attribution vector as a function of class. In this work, we demonstrate that class-dependent methods can \"leak\" information about the selected class, making that class appear more likely than it is. Thus, an end user runs the risk of drawing false conclusions when interpreting an explanation generated by a class-dependent method. In contrast, we introduce \"distribution-aware\" methods, which favor explanations that keep the label's distribution close to its distribution given all features of the input. We introduce SHAP-KL and FastSHAP-KL, two baseline distribution-aware methods that compute Shapley values. Finally, we perform a comprehensive evaluation of seven class-dependent and three distribution-aware methods on three clinical datasets of different high-dimensional data types: images, biosignals, and text.</p>","PeriodicalId":74504,"journal":{"name":"Proceedings of machine learning research","volume":"206 ","pages":"8925-8953"},"PeriodicalIF":0.0,"publicationDate":"2023-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12022845/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144045795","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}