Modern deep learning algorithms tend to optimize an objective metric, such as minimize a cross entropy loss on a training dataset, to be able to learn. The problem is that the single metric is an incomplete description of the real world tasks. The single metric cannot explain why the algorithm learn. When an erroneous happens, the lack of interpretability causes a hardness of understanding and fixing the error. Recently, there are works done to tackle the problem of interpretability to provide insights into neural networks behavior and thought process. The works are important to identify potential bias and to ensure algorithm fairness as well as expected performance.
The use of sophisticated machine learning models for critical decision making is faced with a challenge that these models are often applied as a "black-box". This has led to an increased interest in interpretable machine learning, where post hoc interpretation presents a useful mechanism for generating interpretations of complex learning models. In this paper, we propose a novel approach underpinned by an extended framework of Bayesian networks for generating post hoc interpretations of a black-box predictive model. The framework supports extracting a Bayesian network as an approximation of th…
Episodic Memory Theory for the Mechanistic Interpretation of Recurrent Neural Networks
Manning College of Information and Computer Sciences（曼宁信息与计算机科学学院）；University of Massachusetts（马萨诸塞大学）；Department of Mathematics and Statistics（数学与统计系）
Neural Style Transfer has recently demonstrated very exciting results which catches eyes in both academia and industry. Despite the amazing results, the principle of neural style transfer, especially why the Gram matrices could represent style remains unclear. In this paper, we propose a novel interpretation of neural style transfer by treating it as a domain adaptation problem. Specifically, we theoretically show that matching the Gram matrices of feature maps is equivalent to minimize the Maximum Mean Discrepancy (MMD) with the second order polynomial kernel. Thus, we argue that the essence …
Deep Neural Networks (DNNs) are becoming an important tool in modern computing applications. Accelerating their training is a major challenge and techniques range from distributed algorithms to low-level circuit design. In this survey, we describe the problem from a theoretical perspective, followed by approaches for its parallelization. We present trends in DNN architectures and the resulting implications on parallelization strategies. We then review and model the different types of concurrency in DNNs: from the single operator, through parallelism in network inference and training, to distri…
We revisit fuzzy neural network with a cornerstone notion of generalized hamming distance, which provides a novel and theoretically justified framework to re-interpret many useful neural network techniques in terms of fuzzy logic. In particular, we conjecture and empirically illustrate that, the celebrated batch normalization (BN) technique actually adapts the normalized bias such that it approximates the rightful bias induced by the generalized hamming distance. Once the due bias is enforced analytically, neither the optimization of bias terms nor the sophisticated batch normalization is need…
Auburn University（奥本大学）；Uber AI Labs（优步人工智能实验室）；University of Wyoming（怀俄明大学）
A neuroscience method to understanding the brain is to find and study the preferred stimuli that highly activate an individual cell or groups of cells. Recent advances in machine learning enable a family of methods to synthesize preferred stimuli that cause a neuron in an artificial or biological brain to fire strongly. Those methods are known as Activation Maximization (AM) or Feature Visualization via Optimization. In this chapter, we (1) review existing AM techniques in the literature; (2) discuss a probabilistic interpretation for AM; and (3) review the applications of AM in debugging and …
An increasing number of machine learning models have been deployed in domains with high stakes such as finance and healthcare. Despite their superior performances, many models are black boxes in nature which are hard to explain. There are growing efforts for researchers to develop methods to interpret these black-box models. Post hoc explanations based on perturbations, such as LIME, are widely used approaches to interpret a machine learning model after it has been built. This class of methods has been shown to exhibit large instability, posing serious challenges to the effectiveness of the me…
Baidu Research, Baidu Inc.（百度公司百度研究院）；University of Pennsylvania（宾夕法尼亚大学）；Tsinghua University（清华大学）；University of Oregon（俄勒冈大学）
Deep neural networks have been well-known for their superb handling of various machine learning and artificial intelligence tasks. However, due to their over-parameterized black-box nature, it is often difficult to understand the prediction results of deep models. In recent years, many interpretation tools have been proposed to explain or reveal how deep models make decisions. In this paper, we review this line of research and try to make a comprehensive survey. Specifically, we first introduce and clarify two basic concepts -- interpretations and interpretability -- that people usually get co…
Attention Sink in Transformers: A Survey on Utilization, Interpretation, and Mitigation
Tsinghua University（清华大学）；Meituan LongCat Team（美团LongCat团队）；The University of Hong Kong（香港大学）；University of Michigan（密歇根大学）；Xiamen University（厦门大学）；The Ohio State University（俄亥俄州立大学）；Columbia University（哥伦比亚大学）；Shanghai Artificial Intelligence Laboratory（上海人工智能实验室）；The Hong Kong Polytechnic University（香港理工大学）
Beijing Key Laboratory of Security and Privacy in Intelligent Transportation, Beijing Jiaotong University（北京交通大学智能交通信息安全与隐私保护北京市重点实验室）
Malware is being increasingly threatening and malware detectors based on traditional signature-based analysis are no longer suitable for current malware detection. Recently, the models based on machine learning (ML) are developed for predicting unknown malware variants and saving human strength. However, most of the existing ML models are black-box, which made their pre-diction results undependable, and therefore need further interpretation in order to be effectively deployed in the wild. This paper aims to examine and categorize the existing researches on ML-based malware detector interpretab…
Deep learning can extract rich data representations if provided sufficient quantities of labeled training data. For many tasks however, annotating data has significant costs in terms of time and money, owing to the high standards of subject matter expertise required, for example in medical and geophysical image interpretation tasks. Active Learning can identify the most informative training examples for the interpreter to train, leading to higher efficiency. We propose an Active learning method based on jointly learning representations for supervised and unsupervised tasks. The learned manifol…
Xi’an Jiaotong-Liverpool University（西交利物浦大学）；Duke Kunshan University（昆山杜克大学）；Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；University of Chinese Academy of Sciences（中国科学院大学）
In the last a few decades, deep neural networks have achieved remarkable success in machine learning, computer vision, and pattern recognition. Recent studies however show that neural networks (both shallow and deep) may be easily fooled by certain imperceptibly perturbed input samples called adversarial examples. Such security vulnerability has resulted in a large body of research in recent years because real-world threats could be introduced due to vast applications of neural networks. To address the robustness issue to adversarial examples particularly in pattern recognition, robust adversa…
National Taiwan University（国立台湾大学）；Academia Sinica（中央研究院）
Attribute-aware CF models aims at rating prediction given not only the historical rating from users to items, but also the information associated with users (e.g. age), items (e.g. price), or even ratings (e.g. rating time). This paper surveys works in the past decade developing attribute-aware CF systems, and discovered that mathematically they can be classified into four different categories. We provide the readers not only the high level mathematical interpretation of the existing works in this area but also the mathematical insight for each category of models. Finally we provide in-depth e…
Missouri University of Science and Technology（密苏里科技大学）
This survey samples from the ever-growing family of adaptive resonance theory (ART) neural network models used to perform the three primary machine learning modalities, namely, unsupervised, supervised and reinforcement learning. It comprises a representative list from classic to modern ART models, thereby painting a general picture of the architectures developed by researchers over the past 30 years. The learning dynamics of these ART models are briefly described, and their distinctive characteristics such as code representation, long-term memory and corresponding geometric interpretation are…
TuSimple Inc.（图森未来公司）；Georgia Institute of Technology（佐治亚理工学院）；University of California, Berkeley（加州大学伯克利分校）；California PATH（加州先进交通与高速公路合作伙伴项目）
In this survey, we systematically summarize the current literature on studies that apply reinforcement learning (RL) to the motion planning and control of autonomous vehicles. Many existing contributions can be attributed to the pipeline approach, which consists of many hand-crafted modules, each with a functionality selected for the ease of human interpretation. However, this approach does not automatically guarantee maximal performance due to the lack of a system-level optimization. Therefore, this paper also presents a growing trend of work that falls into the end-to-end approach, which typ…
Machine learning-based medical anomaly detection is an important problem that has been extensively studied. Numerous approaches have been proposed across various medical application domains and we observe several similarities across these distinct applications. Despite this comparability, we observe a lack of structured organisation of these diverse research applications such that their advantages and limitations can be studied. The principal aim of this survey is to provide a thorough theoretical analysis of popular deep learning techniques in medical anomaly detection. In particular, we cont…
