Deep multiple instance selection

Li, Xinchun; Zhan, De-Chuan; Yang, Jia-Qi; Shi, Yi

doi:10.1007/s11432-020-3117-3

Cited by 20 publications

(7 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, the DMIS approach reached a classification accuracy of 0.907 on the MUSK2 dataset while state-of-the-art approaches had 0.836-0.903. 93 Shin et al 95 applied a neural network inversion mechanism 96 in the MIL classification problem and demonstrated that it can significantly improve the KID performance. In the image classification tasks (MNIST, colon cancer, and breast cancer) the approach achieved F1 scores of 0.65, 0.75, and 0.23, while conventional attention reached 0.29, 0.33 and 0.15, respectively.…”

Section: Key Instance Detection Algorithmsmentioning

confidence: 99%

Chemical complexity challenge: Is multi‐instance machine learning a solution?

Zankov,

Madzhidov,

Varnek

et al. 2023

WIREs Comput Mol Sci

View full text Add to dashboard Cite

Molecules are complex dynamic objects that can exist in different molecular forms (conformations, tautomers, stereoisomers, protonation states, etc.) and often it is not known which molecular form is responsible for observed physicochemical and biological properties of a given molecule. This raises the problem of the selection of the correct molecular form for machine learning modeling of target properties. The same problem is common to biological molecules (RNA, DNA, proteins)—long sequences where only key segments, which often cannot be located precisely, are involved in biological functions. Multi‐instance machine learning (MIL) is an efficient approach for solving problems where objects under study cannot be uniquely represented by a single instance, but rather by a set of multiple alternative instances. Multi‐instance learning was formalized in 1997 and motivated by the problem of conformation selection in drug activity prediction tasks. Since then MIL has found a lot of applications in various domains, such as information retrieval, computer vision, signal processing, bankruptcy prediction, and so on. In the given review we describe the MIL framework and its applications to the tasks associated with ambiguity in the representation of small and biological molecules in chemoinformatics and bioinformatics. We have collected examples that demonstrate the advantages of MIL over the traditional single‐instance learning (SIL) approach. Special attention was paid to the ability of MIL models to identify key instances responsible for a modeling property.This article is categorized under: Data Science > Chemoinformatics Data Science > Artificial Intelligence/Machine Learning

show abstract

Section: Key Instance Detection Algorithmsmentioning

confidence: 99%

Chemical complexity challenge: Is multi‐instance machine learning a solution?

Zankov,

Madzhidov,

Varnek

et al. 2023

WIREs Comput Mol Sci

View full text Add to dashboard Cite

show abstract

“…Feature concatenation Feature refinement gcForest [1] Predictions Iterative replacement gcForest S [37] Predictions Confidence screening & Feature screening mdDF [32] Additive predictions Boosting by margin distribution hiDF [31] Interactions Rules extraction by stability s(•, j) j s(x, j) > 0 x…”

Section: Existing Workmentioning

confidence: 99%

“…Since Zhou and Feng [1] propose the original deep forest that significantly improves the performance of tree-based models, several improvements are proposed by designing novel feature concatenations and feature refinements. Pang et al [37] utilize confidence screening and feature screening to refine the concatenated data matrix. The screening method substantially reduces the number of instances that need to be processed and screens out many non-informative features.…”

Section: Existing Improvementsmentioning

confidence: 99%

“…Recently, Lyu et al [32] give a theoretical explanation of the success of a boosting-type feature refinement from the perspective of margin theory. Pang et al [37] prove that the screening-type feature refinement can vary the model complexity from low to high as the number of layers increases in deep forests. However, the role of feature concatenation remains a mystery.…”

Section: Existing Improvementsmentioning

confidence: 99%

“…Lots of work has recently improved the two components of deep forests [ 29,31,32,37 ] and expanded the tree‐based deep models to some specific settings, such as multi‐label learning [ 26 ] , multi‐instance learning [ 38 ] , multi‐modal learning [ 39 ] , semi‐supervised learning [ 27 ] , and crowdsourcing aggregation [ 40 ] . It would be interesting to explore the possibility of exploiting deep forests for rehearsal [ 41 ] .…”

Section: Deep Forestmentioning

confidence: 99%

See 2 more Smart Citations

A Region‐Based Analysis for the Feature Concatenation in Deep Forests

Lyu

CHEN

Zhou

2022

Chinese J of Electronics

View full text Add to dashboard Cite

Deep forest is a tree‐based deep model made up of non‐differentiable modules that are trained without backpropagation. Despite the fact that deep forests have achieved considerable success in a variety of tasks, feature concatenation, as the ingredient for forest representation learning, still lacks theoretical understanding. In this paper, we aim to understand the influence of feature concatenation on predictive performance. To enable such theoretical studies, we present the first mathematical formula of feature concatenation based on the two‐stage structure, which regards the splits along new features and raw features as a region selector and a region classifier respectively. Furthermore, we prove a region‐based generalization bound for feature concatenation, which reveals the trade‐off between Rademacher complexities of the two‐stage structure and the fraction of instances that are correctly classified in the selected region. As a consequence, we show that compared with the prediction‐based feature concatenation (PFC), the advantage of interaction‐based feature concatenation (IFC) is that it obtains more abundant regions through distributed representation and alleviates the overfitting risk in local regions. Experiments confirm the correctness of our theoretical results.

show abstract

Simultaneous instance pooling and bag representation selection approach for multiple-instance learning (MIL) using vision transformer

Waqas,

Tahir,

Author

et al. 2024

Neural Comput & Applic

View full text Add to dashboard Cite

In multiple-instance learning (MIL), the existing bag encoding and attention-based pooling approaches assume that the instances in the bag have no relationship among them. This assumption is unsuited, as the instances in the bags are rarely independent in diverse MIL applications. In contrast, the instance relationship assumption-based techniques incorporate the instance relationship information in the classification process. However, in MIL, the bag composition process is complicated, and it may be possible that instances in one bag are related and instances in another bag are not. In present MIL algorithms, this relationship assumption is not explicitly modeled. The learning algorithm is trained based on one of two relationship assumptions (whether instances in all bags have a relationship or not). Hence, it is essential to model the assumption of instance relationships in the bag classification process. This paper proposes a robust approach that generates vector representation for the bag for both assumptions and the representation selection process to determine whether to consider the instances related or unrelated in the bag classification process. This process helps to determine the essential bag representation vector for every individual bag. The proposed method utilizes attention pooling and vision transformer approaches to generate bag representation vectors. Later, the representation selection subnetwork determines the vector representation essential for bag classification in an end-to-end trainable manner. The generalization abilities of the proposed framework are demonstrated through extensive experiments on several benchmark datasets. The experiments demonstrate that the proposed approach outperforms other state-of-the-art MIL approaches in bag classification.

show abstract

Deep multiple instance selection

Cited by 20 publications

References 25 publications

Chemical complexity challenge: Is multi‐instance machine learning a solution?

Chemical complexity challenge: Is multi‐instance machine learning a solution?

A Region‐Based Analysis for the Feature Concatenation in Deep Forests

Simultaneous instance pooling and bag representation selection approach for multiple-instance learning (MIL) using vision transformer

Contact Info

Product

Resources

About