TALISMAN: Targeted Active Learning for Object Detection with Rare Classes and Slices using Submodular Mutual Information
2. prosince 2022
Řečníci
Suraj Kothawade
Řečník · 0 sledujících
Saikat Ghosh
Řečník · 0 sledujících
Sumit Shekhar
Řečník · 0 sledujících
Rishabh Iyer
Řečník · 0 sledujících
O prezentaci
Deep neural networks based object detectors have shown great success in a variety of domains like autonomous vehicles, biomedical imaging, etc. It is known that their success depends on a large amount of data from the domain of interest. While deep models often perform well in terms of overall accuracy, they often struggle in performance on rare yet critical data slices. For example, data slices like "motorcycle at night" or "bicycle at night" are often rare but very critical slices for self-driving applications and false negatives on such rare slices could result in ill-fated failures and accidents. Active learning (AL) is a well-known paradigm to incrementally and adaptively build training datasets with a human in the loop. However, current AL based acquisition functions are not well-equipped to tackle real-world datasets with rare slices, since they are based on uncertainty scores or global descriptors of the image. We propose TALISMAN, a novel framework for Targeted Active Learning or object detectIon with rare slices using Submodular MutuAl iNformation. Our method uses the submodular mutual information functions instantiated using features of the region of interest (RoI) to efficiently target and acquire data points with rare slices. We evaluate our framework on the standard PASCAL VOC07+12 and BDD100K, a real-world self-driving dataset. We observe that TALISMAN outperforms other methods by in terms of average precision on rare slices, and in terms of mAP.Deep neural networks based object detectors have shown great success in a variety of domains like autonomous vehicles, biomedical imaging, etc. It is known that their success depends on a large amount of data from the domain of interest. While deep models often perform well in terms of overall accuracy, they often struggle in performance on rare yet critical data slices. For example, data slices like "motorcycle at night" or "bicycle at night" are often rare but very critical slices for self-dri…
Organizátor
NeurIPS 2022
Účet · 962 sledujících
Baví vás formát? Nechte SlidesLive zachytit svou akci!
Profesionální natáčení a streamování po celém světě.
Sdílení
Doporučená videa
Prezentace na podobné téma, kategorii nebo přednášejícího
First is Better Than Last for Language Data Influence
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Generating Training Data with Language Models: Towards Zero-Shot Language Understanding
Zhlédnout později
Yu Meng, …
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Chromatic Correlation Clustering, Revisited
Zhlédnout později
Qing Xiu, …
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
A Design-Based Riesz Representation Framework For Randomized Experiments
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Interpolation and Regularization for Causal Learning
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Conditional Moment Alignment for Improved Generalization in Federated Learning
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %