Boosting Meta-Training with Base Class Information for Few-Shot Learning

Few-shot learning, a challenging task in machine learning, aims to learn a classifier adaptable to recognize new, unseen classes with limited labeled examples. Meta-learning has emerged as a prominent framework for few-shot learning. Its training framework is originally a task-level learning method, such as Model-Agnostic Meta-Learning (MAML) and Prototypical Networks. And a recently proposed training paradigm called Meta-Baseline, which consists of sequential pre-training and meta-training stages, gains state-of-the-art performance. However, as a non-end-to-end training method, indicating the meta-training stage can only begin after the completion of pre-training, Meta-Baseline suffers from higher training cost and suboptimal performance due to the inherent conflicts of the two training stages. To address these limitations, we propose an end-to-end training paradigm consisting of two alternative loops. In the outer loop, we calculate cross entropy loss on the entire training set while updating only the final linear layer. In the inner loop, we employ the original meta-learning training mode to calculate the loss and incorporate gradients from the outer loss to guide the parameter updates. This training paradigm not only converges quickly but also outperforms existing baselines, indicating that information from the overall training set and the meta-learning training paradigm could mutually reinforce one another. Moreover, being model-agnostic, our framework achieves significant performance gains, surpassing the baseline systems by approximate 1%.

Your Vision-Language Model Itself Is a Strong Filter: Towards High-Quality Instruction Tuning with Data Selection

Data selection in instruction tuning emerges as a pivotal process for acquiring high-quality data and training instruction-following large language models (LLMs), but it is still a new and unexplored research area for vision-language models (VLMs). Existing data selection approaches on LLMs either rely on single unreliable scores, or use downstream tasks for selection, which is time-consuming and can lead to potential over-fitting on the chosen evaluation datasets. To address this challenge, we introduce a novel dataset selection method, Self-Filter, that utilizes the VLM itself as a filter. This approach is inspired by the observation that VLMs benefit from training with the most challenging instructions. Self-Filter operates in two stages. In the first stage, we devise a scoring network to evaluate the difficulty of training instructions, which is co-trained with the VLM. In the second stage, we use the trained score net to measure the difficulty of each instruction, select the most challenging samples, and penalize similar samples to encourage diversity. Comprehensive experiments on LLaVA and MiniGPT-4 show that Self-Filter can reach better results compared to full data settings with merely about 15% samples, and can achieve superior performance against competitive baselines.

GPT-4 Vision on Medical Image Classification -- A Case Study on COVID-19 Dataset

This technical report delves into the application of GPT-4 Vision (GPT-4V) in the nuanced realm of COVID-19 image classification, leveraging the transformative potential of in-context learning to enhance diagnostic processes.

Rethinking Class Activation Maps for Segmentation: Revealing Semantic Information in Shallow Layers by Reducing Noise

Class activation maps are widely used for explaining deep neural networks. Due to its ability to highlight regions of interest, it has evolved in recent years as a key step in weakly supervised learning. A major limitation to the performance of the class activation maps is the small spatial resolution of the feature maps in the last layer of the convolutional neural network. Therefore, we expect to generate high-resolution feature maps that result in high-quality semantic information. In this paper, we rethink the properties of semantic information in shallow feature maps. We find that the shallow feature maps still have fine-grained non-discriminative features while mixing considerable non-target noise. Furthermore, we propose a simple gradient-based denoising method to filter the noise by truncating the positive gradient. Our proposed scheme can be easily deployed in other CAM-related methods, facilitating these methods to obtain higher-quality class activation maps. We evaluate the proposed approach through a weakly-supervised semantic segmentation task, and a large number of experiments demonstrate the effectiveness of our approach.

CG-fusion CAM: Online segmentation of laser-induced damage on large-aperture optics

Online segmentation of laser-induced damage on large-aperture optics in high-power laser facilities is challenged by complicated damage morphology, uneven illumination and stray light interference. Fully supervised semantic segmentation algorithms have achieved state-of-the-art performance, but rely on plenty of pixel-level labels, which are time-consuming and labor-consuming to produce. LayerCAM, an advanced weakly supervised semantic segmentation algorithm, can generate pixel-accurate results using only image-level labels, but its scattered and partially under-activated class activation regions degrade segmentation performance. In this paper, we propose a weakly supervised semantic segmentation method with Continuous Gradient CAM and its nonlinear multi-scale fusion (CG-fusion CAM). The method redesigns the way of back-propagating gradients and non-linearly activates the multi-scale fused heatmaps to generate more fine-grained class activation maps with appropriate activation degree for different sizes of damage sites. Experiments on our dataset show that the proposed method can achieve segmentation performance comparable to that of fully supervised algorithms.

Effective Medical Code Prediction via Label Internal Alignment

The clinical notes are usually typed into the system by physicians. They are typically required to be marked by standard medical codes, and each code represents a diagnosis or medical treatment procedure. Annotating these notes is time consuming and prone to error. In this paper, we proposed a multi-view attention based Neural network to predict medical codes from clinical texts. Our method incorporates three aspects of information, the semantic context of the clinical text, the relationship among the label (medical codes) space, and the alignment between each pair of a clinical text and medical code. Our method is verified to be effective on the open source dataset. The experimental result shows that our method achieves better performance against the prior state-of-art on multiple metrics.

SuperScaler: Supporting Flexible DNN Parallelization via a Unified Abstraction

With the growing model size, deep neural networks (DNN) are increasingly trained over massive GPU accelerators, which demands a proper parallelization plan that transforms a DNN model into fine-grained tasks and then schedules them to GPUs for execution. Due to the large search space, the contemporary parallelization plan generators often rely on empirical rules that couple transformation and scheduling, and fall short in exploring more flexible schedules that yield better memory usage and compute efficiency. This tension can be exacerbated by the emerging models with increasing complexity in their structure and model size. SuperScaler is a system that facilitates the design and generation of highly flexible parallelization plans. It formulates the plan design and generation into three sequential phases explicitly: model transformation, space-time scheduling, and data dependency preserving. Such a principled approach decouples multiple seemingly intertwined factors and enables the composition of highly flexible parallelization plans. As a result, SuperScaler can not only generate empirical parallelization plans, but also construct new plans that achieve up to 3.5X speedup compared to state-of-the-art solutions like DeepSpeed, Megatron and Alpa, for emerging DNN models like Swin-Transformer and AlphaFold2, as well as well-optimized models like GPT-3.

Fast optical refocusing through multimode fiber bend using Cake-Cutting Hadamard encoding algorithm to improve robustness

Multimode fibres offer the advantages of high resolution and miniaturization over single mode fibers in the field of optical imaging. However, multimode fibre's imaging is susceptible to perturbations of MMF that can lead to secondary spatial distortions in the transmitted image. Perturbations include random disturbances in the fiber as well as environmental noise. Here, we exploit the fast focusing capability of the Cake-Cutting Hadamard coding algorithm to counteract the effects of perturbations and improve the system's robustness. Simulation shows that it can approach the theoretical enhancement at 2000 measurements. Experimental results show that the algorithm can help the system to refocus in a short time when MMFs are perturbed. This research will further contribute to using multimode fibres in medicine, communication, and detection.

Generating Pseudo-labels Adaptively for Few-shot Model-Agnostic Meta-Learning

Model-Agnostic Meta-Learning (MAML) is a famous few-shot learning method that has inspired many follow-up efforts, such as ANIL and BOIL. However, as an inductive method, MAML is unable to fully utilize the information of query set, limiting its potential of gaining higher generality. To address this issue, we propose a simple yet effective method that generates psuedo-labels adaptively and could boost the performance of the MAML family. The proposed methods, dubbed Generative Pseudo-label based MAML (GP-MAML), GP-ANIL and GP-BOIL, leverage statistics of the query set to improve the performance on new tasks. Specifically, we adaptively add pseudo labels and pick samples from the query set, then re-train the model using the picked query samples together with the support set. The GP series can also use information from the pseudo query set to re-train the network during the meta-testing. While some transductive methods, such as Transductive Propagation Network (TPN), struggle to achieve this goal.

Training Neural Networks for Solving 1-D Optimal Piecewise Linear Approximation

Recently, the interpretability of deep learning has attracted a lot of attention. A plethora of methods have attempted to explain neural networks by feature visualization, saliency maps, model distillation, and so on. However, it is hard for these methods to reveal the intrinsic properties of neural networks. In this work, we studied the 1-D optimal piecewise linear approximation (PWLA) problem, and associated it with a designed neural network, named lattice neural network (LNN). We asked four essential questions as following: (1) What are the characters of the optimal solution of the PWLA problem? (2) Can an LNN converge to the global optimum? (3) Can an LNN converge to the local optimum? (4) Can an LNN solve the PWLA problem? Our main contributions are that we propose the theorems to characterize the optimal solution of the PWLA problem and present the LNN method for solving it. We evaluated the proposed LNNs on approximation tasks, forged an empirical method to improve the performance of LNNs. The experiments verified that our LNN method is competitive with the start-of-the-art method.