The paper "TADAM: Task dependent adaptive metric for improved few-shot learning" by Boris N. Oreshkin addresses the challenge of few-shot learning, where models must generalize from a small number of examples. The authors identify that metric scaling and task conditioning are crucial for enhancing the performance of few-shot algorithms. They propose a method to scale the similarity metric, which can improve accuracy by up to 14% on the mini-Imagenet 5-way 5-shot classification task. Additionally, they introduce a task-dependent metric space by conditioning the feature extractor on the specific task, using a task encoding network to extract a task representation. The authors also propose an end-to-end optimization procedure based on auxiliary task co-training to learn this task-dependent metric space. The resulting model achieves state-of-the-art performance on mini-Imagenet and shows significant improvements over existing methods on a new few-shot dataset based on CIFAR100. The code for the proposed method is publicly available.The paper "TADAM: Task dependent adaptive metric for improved few-shot learning" by Boris N. Oreshkin addresses the challenge of few-shot learning, where models must generalize from a small number of examples. The authors identify that metric scaling and task conditioning are crucial for enhancing the performance of few-shot algorithms. They propose a method to scale the similarity metric, which can improve accuracy by up to 14% on the mini-Imagenet 5-way 5-shot classification task. Additionally, they introduce a task-dependent metric space by conditioning the feature extractor on the specific task, using a task encoding network to extract a task representation. The authors also propose an end-to-end optimization procedure based on auxiliary task co-training to learn this task-dependent metric space. The resulting model achieves state-of-the-art performance on mini-Imagenet and shows significant improvements over existing methods on a new few-shot dataset based on CIFAR100. The code for the proposed method is publicly available.