Multi-Instance Learning : A Survey

By Zhi-Hua Zhou. 2004

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Learning Theory for Distribution Regression (2014, methodology): “The MIL approach has been applied in several domains; see the reviews from Babenko (2004); Zhou (2004); Foulds and Frank (2010); Amores (2013).”
A Closer Look at Weak Label Learning for Audio Events (2018, methodology): “[37] proposed scalable MIL methods for audio event classiﬁcation using weak labels.”
On classification with bags, groups and sets (2014, background): “More extensive surveys of MIL assumptions and classifiers can be found in [59, 13, 1].”
W-TALC: Weakly-supervised Temporal Activity Localization and Classification (2018, methodology): “As discussed in Section 1, the weakly-supervised activity localization and classification problem as addressed in this paper can be directly mapped to the problem of Multiple Instance Learning (MIL) [70].”
Multiple Structure-View Learning for Graph Classification (2018, background): “This representation and learning complication is known as multi-instance learning [8].”
Identifying Multi-instance Outliers (2010, unknown): “We find that such conclusion is consistent with the summarization in [13] which shows that the error classification rate of Musk1 is usually smaller than that of Musk2 under the same multi-instance learning method (please see Table 2 in [13]).”
Multiple instance learning: A survey of problem characteristics and applications (2016, background): “It is likely that instances of the same molecule are similar to some extent, while being different from other molecules [13].”
Completeness Modeling and Context Separation for Weakly Supervised Temporal Action Localization (2019, methodology): “We compute its cross-entropy with the ground truth, i.e., the standard MIL loss, which is minimized along with the diversity loss to learn the network parameters.”
Unbox the black-box for the medical explainable AI via multi-modal and multi-centre data fusion: A mini-review, two showcases and beyond (2021, methodology): “Therefore, we adopt a section based strategy to handle this problem and fit this into a Multiple Instance Learning (MIL) framework [166].”