Deep learning has been widely used in data-intensive applications. However, training a deep neural network often requires a large data set. When there is not enough data available for training, the performance of deep learning models is even worse than that of shallow networks. It has been proved that few-shot learning can generalize to new tasks with few training samples. Fine-tuning of a deep model is simple and effective few-shot learning method. However, how to fine-tune deep learning models (fine-tune convolution layer or BN layer?) still lack deep investigation. Hence, we study how to fine-tune deep models through experimental comparison in this paper. Furthermore, the weight of the models is analyzed to verify the feasibility of the fine-tuning method.
University of Texas at Austin（德克萨斯大学奥斯汀分校）；University of Colorado, Boulder（科罗拉多大学博尔德分校）；Adobe Research（奥多比研究院）
Generalized few-shot semantic segmentation was introduced to move beyond only evaluating few-shot segmentation models on novel classes to include testing their ability to remember base classes. While the current state-of-the-art approach is based on meta-learning, it performs poorly and saturates in learning after observing only a few shots. We propose the first fine-tuning solution, and demonstrate that it addresses the saturation problem while achieving state-of-the-art results on two datasets, PASCAL-5i and COCO-20i. We also show that it outperforms existing methods, whether fine-tuning mul…
Carnegie Mellon University（卡内基梅隆大学）；Stanford University（斯坦福大学）；Bosch Center for AI（博世人工智能中心）
Finetuning image-text models such as CLIP achieves state-of-the-art accuracies on a variety of benchmarks. However, recent works like WiseFT (Wortsman et al., 2021) and LP-FT (Kumar et al., 2022) have shown that even subtle differences in the finetuning process can lead to surprisingly large differences in the final performance, both for in-distribution (ID) and out-of-distribution (OOD) data. In this work, we show that a natural and simple approach of mimicking contrastive pretraining consistently outperforms alternative finetuning approaches. Specifically, we cast downstream class labels as …
Samsung AI Center Cambridge（三星剑桥人工智能中心）；University of Edinburgh（爱丁堡大学）
Few-shot learning (FSL) is an important and topical problem in computer vision that has motivated extensive research into numerous methods spanning from sophisticated meta-learning methods to simple transfer learning baselines. We seek to push the limits of a simple-but-effective pipeline for more realistic and practical settings of few-shot image classification. To this end, we explore few-shot learning from the perspective of neural network architecture, as well as a three stage pipeline of network updates under different data supplies, where unsupervised external data is considered for pre-…
Prototypical networks have been shown to perform well at few-shot learning tasks in computer vision. Yet these networks struggle when classes are very similar to each other (fine-grain classification) and currently have no way of taking into account prior knowledge (through the use of tabular data). Using a spherical latent space to encode prototypes, we can achieve few-shot fine-grain classification by maximally separating the classes while incorporating domain knowledge as informative priors. We describe how to construct a hypersphere of prototypes that embed a-priori domain information, and…
In this paper, we tackle the new Cross-Domain Few-Shot Learning benchmark proposed by the CVPR 2020 Challenge. To this end, we build upon state-of-the-art methods in domain adaptation and few-shot learning to create a system that can be trained to perform both tasks. Inspired by the need to create models designed to be fine-tuned, we explore the integration of transfer-learning (fine-tuning) with meta-learning algorithms, to train a network that has specific layers that are designed to be adapted at a later fine-tuning stage. To do so, we modify the episodic training process to include a first…
Huazhong University of Science and Technology（华中科技大学）
Different from fine-tuning models pre-trained on a large-scale dataset of preset classes, class-incremental learning (CIL) aims to recognize novel classes over time without forgetting pre-trained classes. However, a given model will be challenged by test images with finer-grained classes, e.g., a basenji is at most recognized as a dog. Such images form a new training set (i.e., support set) so that the incremental model is hoped to recognize a basenji (i.e., query) as a basenji next time. This paper formulates such a hybrid natural problem of coarse-to-fine few-shot (C2FS) recognition as a CIL…
Few-shot learning is the process of learning novel classes using only a few examples and it remains a challenging task in machine learning. Many sophisticated few-shot learning algorithms have been proposed based on the notion that networks can easily overfit to novel examples if they are simply fine-tuned using only a few examples. In this study, we show that in the commonly used low-resolution mini-ImageNet dataset, the fine-tuning method achieves higher accuracy than common few-shot learning algorithms in the 1-shot task and nearly the same accuracy as that of the state-of-the-art algorithm…
Few-shot learning (FSL) aims to learn novel visual categories from very few samples, which is a challenging problem in real-world applications. Many methods of few-shot classification work well on general images to learn global representation. However, they can not deal with fine-grained categories well at the same time due to a lack of subtle and local information. We argue that localization is an efficient approach because it directly provides the discriminative regions, which is critical for both general classification and fine-grained classification in a low data regime. In this paper, we …
Few-shot learning aims at rapidly adapting to novel categories with only a handful of samples at test time, which has been predominantly tackled with the idea of meta-learning. However, meta-learning approaches essentially learn across a variety of few-shot tasks and thus still require large-scale training data with fine-grained supervision to derive a generalized model, thereby involving prohibitive annotation cost. In this paper, we advance the few-shot classification paradigm towards a more challenging scenario, i.e., cross-granularity few-shot classification, where the model observes only …
UC Irvine（加州大学欧文分校）；University of Edinburgh（爱丁堡大学）；UC Berkeley（加州大学伯克利分校）；Facebook AI Research（脸书人工智能研究院）；University College London（伦敦大学学院）
Prompting language models (LMs) with training examples and task descriptions has been seen as critical to recent successes in few-shot learning. In this work, we show that finetuning LMs in the few-shot setting can considerably reduce the need for prompt engineering. In fact, one can use null prompts, prompts that contain neither task-specific templates nor training examples, and achieve competitive accuracy to manually-tuned prompts across a wide range of tasks. While finetuning LMs does introduce new parameters for each downstream task, we show that this memory overhead can be substantially …
Multilingual contextual embeddings have demonstrated state-of-the-art performance in zero-shot cross-lingual transfer learning, where multilingual BERT is fine-tuned on one source language and evaluated on a different target language. However, published results for mBERT zero-shot accuracy vary as much as 17 points on the MLDoc classification task across four papers. We show that the standard practice of using English dev accuracy for model selection in the zero-shot setting makes it difficult to obtain reproducible results on the MLDoc and XNLI tasks. English dev accuracy is often uncorrelate…
Princeton University（普林斯顿大学）；Massachusetts Institute of Technology（麻省理工学院）
The recent GPT-3 model (Brown et al., 2020) achieves remarkable few-shot performance solely by leveraging a natural-language prompt and a few task demonstrations as input context. Inspired by their findings, we study few-shot learning in a more practical scenario, where we use smaller language models for which fine-tuning is computationally efficient. We present LM-BFF--better few-shot fine-tuning of language models--a suite of simple and complementary techniques for fine-tuning language models on a small number of annotated examples. Our approach includes (1) prompt-based fine-tuning together…
Technische Universitat Berlin（柏林工业大学）；Bernstein Center for Computational Neuroscience Berlin（柏林伯恩斯坦计算神经科学中心）
In transfer learning, only the last part of the networks - the so-called head - is often fine-tuned. Representation similarity analysis shows that the most significant change still occurs in the head even if all weights are updatable. However, recent results from few-shot learning have shown that representation change in the early layers, which are mostly convolutional, is beneficial, especially in the case of cross-domain adaption. In our paper, we find out whether that also holds true for transfer learning. In addition, we analyze the change of representation in transfer learning, both durin…
NeuralMind；University of Campinas (Unicamp)（坎皮纳斯大学）；Federal University of Goiás (UFG)（戈亚斯联邦大学）；University of Waterloo（滑铁卢大学）
There has been mounting evidence that pretrained language models fine-tuned on large and diverse supervised datasets can transfer well to a variety of out-of-domain tasks. In this work, we investigate this transfer ability to the legal domain. For that, we participated in the legal case entailment task of COLIEE 2021, in which we use such models with no adaptations to the target domain. Our submissions achieved the highest scores, surpassing the second-best team by more than six percentage points. Our experiments confirm a counter-intuitive result in the new paradigm of pretrained language mod…
University of Washington（华盛顿大学）；OpenAI；Columbia University（哥伦比亚大学）；Google Research（谷歌研究院）；Allen Institute for Artificial Intelligence（艾伦人工智能研究所）；Toyota Research Institute（丰田研究院）
Large pre-trained models such as CLIP or ALIGN offer consistent accuracy across a range of data distributions when performing zero-shot inference (i.e., without fine-tuning on a specific dataset). Although existing fine-tuning methods substantially improve accuracy on a given target distribution, they often reduce robustness to distribution shifts. We address this tension by introducing a simple and effective method for improving robustness while fine-tuning: ensembling the weights of the zero-shot and fine-tuned models (WiSE-FT). Compared to standard fine-tuning, WiSE-FT provides large accura…
University of Illinois at Chicago（伊利诺伊大学芝加哥分校）；Adobe Research（Adobe 研究院）
In this work, we focus on a more challenging few-shot intent detection scenario where many intents are fine-grained and semantically similar. We present a simple yet effective few-shot intent detection schema via contrastive pre-training and fine-tuning. Specifically, we first conduct self-supervised contrastive pre-training on collected intent datasets, which implicitly learns to discriminate semantically similar utterances without using any labels. We then perform few-shot intent detection together with supervised contrastive learning, which explicitly pulls utterances from the same intent c…
Small or Large? Zero-Shot or Finetuned? Guiding Language Model Choice for Specialized Applications in Healthcare
British Columbia Cancer Registry Provincial Health Services Authority（不列颠哥伦比亚省癌症登记处省级卫生服务局）；Data Science Institute University of British Columbia（数据科学研究所不列颠哥伦比亚大学）
Pretrained language models demonstrate strong performance in most NLP tasks when fine-tuned on small task-specific datasets. Hence, these autoregressive models constitute ideal agents to operate in text-based environments where language understanding and generative capabilities are essential. Nonetheless, collecting expert demonstrations in such environments is a time-consuming endeavour. We introduce a two-stage procedure to learn from a small set of demonstrations and further improve by interacting with an environment. We show that language models fine-tuned with only 1.2% of the expert demo…
Warsaw University of Technology（华沙理工大学）；NASK - National Research Institute（NASK国家研究所）
This paper presents an analysis regarding an influence of the Distance Metric Learning (DML) loss functions on the supervised fine-tuning of the language models for classification tasks. We experimented with known datasets from SentEval Transfer Tasks. Our experiments show that applying the DML loss function can increase performance on downstream classification tasks of RoBERTa-large models in few-shot scenarios. Models fine-tuned with the use of SoftTriple loss can achieve better results than models with a standard categorical cross-entropy loss function by about 2.89 percentage points from 0…
