FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2948657385> ?p ?o ?g. }

• W2948657385 endingPage "2846" @default.
• W2948657385 startingPage "2832" @default.
• W2948657385 abstract "Class imbalance is a challenging problem in many classification tasks. It induces biased classification results for minority classes that contain less training samples than others. Most existing approaches aim to remedy the imbalanced number of instances among categories by resampling the majority and minority classes accordingly. However, the imbalanced level of difficulty of recognizing different categories is also crucial, especially for distinguishing samples with many classes. For example, in the task of clinical skin disease recognition, several rare diseases have a small number of training samples, but they are easy to diagnose because of their distinct visual properties. On the other hand, some common skin diseases, e.g., eczema, are hard to recognize due to the lack of special symptoms. To address this problem, we propose a self-paced balance learning (SPBL) algorithm in this paper. Specifically, we introduce a comprehensive metric termed the complexity of image category that is a combination of both sample number and recognition difficulty. First, the complexity is initialized using the model of the first pace, where the pace indicates one iteration in the self-paced learning paradigm. We then assign each class a penalty weight that is larger for more complex categories and smaller for easier ones, after which the curriculum is reconstructed by rearranging the training samples. Consequently, the model can iteratively learn discriminative representations via balancing the complexity in each pace. Experimental results on the SD-198 and SD-260 benchmark data sets demonstrate that the proposed SPBL algorithm performs favorably against the state-of-the-art methods. We also demonstrate the effectiveness of the SPBL algorithm's generalization capacity on various tasks, such as indoor scene image recognition and object classification." @default.
• W2948657385 created "2019-06-14" @default.
• W2948657385 creator A5037131575 @default.
• W2948657385 creator A5068129380 @default.
• W2948657385 creator A5072117142 @default.
• W2948657385 creator A5082498609 @default.
• W2948657385 creator A5085119493 @default.
• W2948657385 creator A5086664647 @default.
• W2948657385 creator A5089409678 @default.
• W2948657385 date "2020-08-01" @default.
• W2948657385 modified "2023-10-14" @default.
• W2948657385 title "Self-Paced Balance Learning for Clinical Skin Disease Recognition" @default.
• W2948657385 cites W1500568346 @default.
• W2948657385 cites W1520065212 @default.
• W2948657385 cites W1525219522 @default.
• W2948657385 cites W1551909886 @default.
• W2948657385 cites W1563938718 @default.
• W2948657385 cites W1597870463 @default.
• W2948657385 cites W1912982817 @default.
• W2948657385 cites W1983031733 @default.
• W2948657385 cites W1984712439 @default.
• W2948657385 cites W1992396639 @default.
• W2948657385 cites W1995137594 @default.
• W2948657385 cites W1999318832 @default.
• W2948657385 cites W2011376672 @default.
• W2948657385 cites W2022477494 @default.
• W2948657385 cites W2023639956 @default.
• W2948657385 cites W2024046085 @default.
• W2948657385 cites W2039051437 @default.
• W2948657385 cites W2040010062 @default.
• W2948657385 cites W2040181375 @default.
• W2948657385 cites W2049750909 @default.
• W2948657385 cites W2061119986 @default.
• W2948657385 cites W2072615238 @default.
• W2948657385 cites W2086075853 @default.
• W2948657385 cites W2087787741 @default.
• W2948657385 cites W2088059023 @default.
• W2948657385 cites W2096235960 @default.
• W2948657385 cites W2096945460 @default.
• W2948657385 cites W2098576843 @default.
• W2948657385 cites W2099454382 @default.
• W2948657385 cites W2103614420 @default.
• W2948657385 cites W2104167780 @default.
• W2948657385 cites W2106479238 @default.
• W2948657385 cites W2107327607 @default.
• W2948657385 cites W2108598243 @default.
• W2948657385 cites W2112796928 @default.
• W2948657385 cites W2118978333 @default.
• W2948657385 cites W2119191234 @default.
• W2948657385 cites W2124685890 @default.
• W2948657385 cites W2126734246 @default.
• W2948657385 cites W2132791018 @default.
• W2948657385 cites W2135493362 @default.
• W2948657385 cites W2136256517 @default.
• W2948657385 cites W2142713577 @default.
• W2948657385 cites W2148143831 @default.
• W2948657385 cites W2152161678 @default.
• W2948657385 cites W2164330572 @default.
• W2948657385 cites W2165072487 @default.
• W2948657385 cites W2165430451 @default.
• W2948657385 cites W2166049352 @default.
• W2948657385 cites W2167464971 @default.
• W2948657385 cites W2194775991 @default.
• W2948657385 cites W2296073425 @default.
• W2948657385 cites W2344580970 @default.
• W2948657385 cites W2345010043 @default.
• W2948657385 cites W2346062110 @default.
• W2948657385 cites W2358876993 @default.
• W2948657385 cites W2426942631 @default.
• W2948657385 cites W2490420619 @default.
• W2948657385 cites W2519210008 @default.
• W2948657385 cites W2538394767 @default.
• W2948657385 cites W2564782580 @default.
• W2948657385 cites W2575032143 @default.
• W2948657385 cites W2581082771 @default.
• W2948657385 cites W2615981376 @default.
• W2948657385 cites W2620239873 @default.
• W2948657385 cites W2763619424 @default.
• W2948657385 cites W2765838470 @default.
• W2948657385 cites W2766742395 @default.
• W2948657385 cites W2772315375 @default.
• W2948657385 cites W2775447965 @default.
• W2948657385 cites W2798277818 @default.
• W2948657385 cites W2810809861 @default.
• W2948657385 cites W2884115684 @default.
• W2948657385 cites W2888527098 @default.
• W2948657385 cites W2889081631 @default.
• W2948657385 cites W2900028034 @default.
• W2948657385 cites W2923501483 @default.
• W2948657385 cites W2963702891 @default.
• W2948657385 cites W2964050365 @default.
• W2948657385 cites W4212883601 @default.
• W2948657385 cites W4244952642 @default.
• W2948657385 doi "https://doi.org/10.1109/tnnls.2019.2917524" @default.
• W2948657385 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/31199274" @default.
• W2948657385 hasPublicationYear "2020" @default.
• W2948657385 type Work @default.
• W2948657385 sameAs 2948657385 @default.

• next