Lifeng Gu

Lifeng Gu

As interpretability gains attention in machine learning, there is a growing need for reliable models that fully explain representation content. We propose a mutual information (MI)-based method that decomposes neural network representations into three exhaustive components: total mutual information, decision-related information, and redundant information. This theoretically complete framework captures the entire input-representation relationship, surpassing partial explanations like those from Grad-CAM. Using two lightweight modules integrated into architectures such as CNNs and Transformers,we estimate these components and demonstrate their interpretive power through visualizations on ResNet and prototype network applied to image classification and few-shot learning tasks. Our approach is distinguished by three key features: 1. Rooted in mutual information theory, it delivers a thorough and theoretically grounded interpretation, surpassing the scope of existing interpretability methods. 2. Unlike conventional methods that focus on explaining decisions, our approach centers on interpreting representations. 3. It seamlessly integrates into pre-existing network architectures, requiring only fine-tuning of the inserted modules.

Lifeng Gu

Most existing metric learning methods focus on learning a similarity or distance measure relying on similar and dissimilar relations between sample pairs. However, pairs of samples cannot be simply identified as similar or dissimilar in many real-world applications, e.g., multi-label learning, label distribution learning. To this end, relation alignment metric learning (RAML) framework is proposed to handle the metric learning problem in those scenarios. But RAML learn a linear metric, which can't model complex datasets. Combining with deep learning and RAML framework, we propose a hierarchical relationship alignment metric leaning model HRAML, which uses the concept of relationship alignment to model metric learning problems under multiple learning tasks, and makes full use of the consistency between the sample pair relationship in the feature space and the sample pair relationship in the label space. Further we organize several experiment divided by learning tasks, and verified the better performance of HRAML against many popular methods and RAML framework.

Lifeng Gu

In recent years, representation learning has become the research focus of the machine learning community. Large-scale neural networks are a crucial step toward achieving general intelligence, with their success largely attributed to their ability to learn abstract representations of data. Several learning fields are actively discussing how to learn representations, yet there is a lack of a unified perspective. We convert the representation learning problem under different tasks into a ranking problem. By adopting the ranking problem as a unified perspective, representation learning tasks can be solved in a unified manner by optimizing the ranking loss. Experiments under various learning tasks, such as classification, retrieval, multi-label learning, and regression, prove the superiority of the representation learning by ranking framework. Furthermore, experiments under self-supervised learning tasks demonstrate the significant advantage of the ranking framework in processing unsupervised training data, with data augmentation techniques further enhancing its performance.

Lifeng Gu

Most existing metric learning methods focus on learning a similarity or distance measure relying on similar and dissimilar relations between sample pairs. However, pairs of samples cannot be simply identified as similar or dissimilar in many real-world applications, e.g., multi-label learning, label distribution learning. To this end, relation alignment metric learning (RAML) framework is proposed to handle the metric learning problem in those scenarios. But RAML framework uses SVR solvers for optimization. It can't learn positive semidefinite distance metric which is necessary in metric learning. In this paper, we propose two methds to overcame the weakness. Further, We carry out several experiments on the single-label classification, multi-label classification, label distribution learning to demonstrate the new methods achieves favorable performance against RAML framework.