Applied AI letters最新文献

The goal of continual learning (CL) is to learn a sequence of tasks without suffering from the phenomenon of catastrophic forgetting. Previous work has shown that leveraging memory in the form of a replay buffer can reduce performance degradation on prior tasks. We hypothesize that forgetting can be further reduced when the model is encouraged to remember the evidence for previously made decisions. As a first step towards exploring this hypothesis, we propose a simple novel training paradigm, called Remembering for the Right Reasons (RRR), that additionally stores visual model explanations for each example in the buffer and ensures the model has “the right reasons” for its predictions by encouraging its explanations to remain consistent with those used to make decisions at training time. Without this constraint, there is a drift in explanations and increase in forgetting as conventional continual learning algorithms learn new tasks. We demonstrate how RRR can be easily added to any memory or regularization-based approach and results in reduced forgetting, and more importantly, improved model explanations. We have evaluated our approach in the standard and few-shot settings and observed a consistent improvement across various CL approaches using different architectures and techniques to generate model explanations and demonstrated our approach showing a promising connection between explainability and continual learning. Our code is available at https://github.com/SaynaEbrahimi/Remembering-for-the-Right-Reasons.

We present a novel augmented reality (AR) interface to provide effective means to diagnose a robot's erroneous behaviors, endow it with new skills, and patch its knowledge structure represented by an And-Or-Graph (AOG). Specifically, an AOG representation of opening medicine bottles is learned from human demonstration and yields a hierarchical structure that captures the spatiotemporal compositional nature of the given task, which is highly interpretable for the users. Through a series of psychological experiments, we demonstrate that the explanations of a robotic system, inherited from and produced by the AOG, can better foster human trust compared to other forms of explanations. Moreover, by visualizing the knowledge structure and robot states, the AR interface allows human users to intuitively understand what the robot knows, supervise the robot's task planner, and interactively teach the robot with new actions. Together, users can quickly identify the reasons for failures and conveniently patch the current knowledge structure to prevent future errors. This capability demonstrates the interpretability of our knowledge representation and the new forms of interactions afforded by the proposed AR interface.

Quantifying the value of explanations in a human-in-the-loop (HITL) system is difficult. Previous methods either measure explanation-specific values that do not correspond to user tasks and needs or poll users on how useful they find the explanations to be. In this work, we quantify how much explanations help the user through a utility-based paradigm that measures change in task performance when using explanations vs not. Our chosen task is content-based image retrieval (CBIR), which has well-established baselines and performance metrics independent of explainability. We extend an existing HITL image retrieval system that incorporates user feedback with similarity-based saliency maps (SBSM) that indicate to the user which parts of the retrieved images are most similar to the query image. The system helps the user understand what it is paying attention to through saliency maps, and the user helps the system understand their goal through saliency-guided relevance feedback. Using the MS-COCO dataset, a standard object detection and segmentation dataset, we conducted extensive, crowd-sourced experiments validating that SBSM improves interactive image retrieval. Although the performance increase is modest in the general case, in more difficult cases such as cluttered scenes, using explanations yields an 6.5% increase in accuracy. To the best of our knowledge, this is the first large-scale user study showing that visual saliency map explanations improve performance on a real-world, interactive task. Our utility-based evaluation paradigm is general and potentially applicable to any task for which explainability can be incorporated.

We study a user-guided approach for producing global explanations of deep networks for image recognition. The global explanations are produced with respect to a test data set and give the overall frequency of different “recognition reasons” across the data. Each reason corresponds to a small number of the most significant human-recognizable visual concepts used by the network. The key challenge is that the visual concepts cannot be predetermined and those concepts will often not correspond to existing vocabulary or have labeled data sets. We address this issue via an interactive-naming interface, which allows users to freely cluster significant image regions in the data into visually similar concepts. Our main contribution is a user study on two visual recognition tasks. The results show that the participants were able to produce a small number of visual concepts sufficient for explanation and that there was significant agreement among the concepts, and hence global explanations, produced by different participants.

Recent advances in artificial intelligence (AI), driven mainly by deep neural networks, have yielded remarkable progress in fields, such as computer vision, natural language processing, and reinforcement learning. Despite these successes, the inability to predict how AI systems will behave “in the wild” impacts almost all stages of planning and deployment, including research and development, verification and validation, and user trust and acceptance. The field of explainable artificial intelligence (XAI) seeks to develop techniques enabling AI algorithms to generate explanations of their results; generally these are human-interpretable representations or visualizations that are meant to “explain” how the system produced its outputs. We introduce the Explainable AI Toolkit (XAITK), a DARPA-sponsored effort that builds on results from the 4-year DARPA XAI program. The XAITK has two goals: (a) to consolidate research results from DARPA XAI into a single publicly accessible repository; and (b) to identify operationally relevant capabilities developed on DARPA XAI and assist in their transition to interested partners. We first describe the XAITK website and associated capabilities. These place the research results from DARPA XAI in the wider context of general research in the field of XAI, and include performer contributions of code, data, publications, and reports. We then describe the XAITK analytics and autonomy software frameworks. These are Python-based frameworks focused on particular XAI domains, and designed to provide a single integration endpoint for multiple algorithm implementations from across DARPA XAI. Each framework generalizes APIs for system-level data and control while providing a plugin interface for existing and future algorithm implementations. The XAITK project can be followed at: https://xaitk.org.

In complex inferential tasks like question answering, machine learning models must confront two challenges: the need to implement a compositional reasoning process, and, in many applications, the need for this reasoning process to be interpretable to assist users in both development and prediction. Existing models designed to produce interpretable traces of their decision-making process typically require these traces to be supervised at training time. In this paper, we present a novel neural modular approach that performs compositional reasoning by automatically inducing a desired subtask decomposition without relying on strong supervision. Our model allows linking different reasoning tasks through shared modules that handle common routines across tasks. Experiments show that the model is more interpretable to human evaluators compared to other state-of-the-art models: users can better understand the model's underlying reasoning procedure and predict when it will succeed or fail based on observing its intermediate outputs.

Neural network architectures are achieving superhuman performance on an expanding range of tasks. To effectively and safely deploy these systems, their decision-making must be understandable to a wide range of stakeholders. Methods to explain artificial intelligence (AI) have been proposed to answer this challenge, but a lack of theory impedes the development of systematic abstractions, which are necessary for cumulative knowledge gains. We propose Bayesian Teaching as a framework for unifying explainable AI (XAI) by integrating machine learning and human learning. Bayesian Teaching formalizes explanation as a communication act of an explainer to shift the beliefs of an explainee. This formalization decomposes a wide range of XAI methods into four components: (a) the target inference, (b) the explanation, (c) the explainee model, and (d) the explainer model. The abstraction afforded by Bayesian Teaching to decompose XAI methods elucidates the invariances among them. The decomposition of XAI systems enables modular validation, as each of the first three components listed can be tested semi-independently. This decomposition also promotes generalization through recombination of components from different XAI systems, which facilitates the generation of novel variants. These new variants need not be evaluated one by one provided that each component has been validated, leading to an exponential decrease in development time. Finally, by making the goal of explanation explicit, Bayesian Teaching helps developers to assess how suitable an XAI system is for its intended real-world use case. Thus, Bayesian Teaching provides a theoretical framework that encourages systematic, scientific investigation of XAI.

“In what circumstances would you want this AI to make decisions on your behalf?” We have been investigating how to enable a user of an Artificial Intelligence-powered system to answer questions like this through a series of empirical studies, a group of which we summarize here. We began the series by (a) comparing four explanation configurations of saliency explanations and/or reward explanations. From this study we learned that, although some configurations had significant strengths, no one configuration was a clear “winner.” This result led us to hypothesize that one reason for the low success rates Explainable AI (XAI) research has in enabling users to create a coherent mental model is that the AI itself does not have a coherent model. This hypothesis led us to (b) build a model-based agent, to compare explaining it with explaining a model-free agent. Our results were encouraging, but we then realized that participants' cognitive energy was being sapped by having to create not only a mental model, but also a process by which to create that mental model. This realization led us to (c) create such a process (which we term After-Action Review for AI or “AAR/AI”) for them, integrate it into the explanation environment, and compare participants' success with AAR/AI scaffolding vs without it. Our AAR/AI studies' results showed that AAR/AI participants were more effective assessing the AI than non-AAR/AI participants, with significantly better precision and significantly better recall at finding the AI's reasoning flaws.

With the increasing deployment of smart buildings and infrastructure, supervisory control and data acquisition (SCADA) devices and the underlying IT network have become essential elements for the proper operations of these highly complex systems. Of course, with the increase in automation and the proliferation of SCADA devices, a corresponding increase in surface area of attack on critical infrastructure has increased. Understanding device behaviors in terms of known and understood or potentially qualified activities vs unknown and potentially nefarious activities in near-real time is a key component of any security solution. In this paper, we investigate the challenges with building robust machine learning models to identify unknowns purely from network traffic both inside and outside firewalls, starting with missing or inconsistent labels across sites, feature engineering and learning, temporal dependencies and analysis, and training data quality (including small sample sizes) for both shallow and deep learning methods. To demonstrate these challenges and the capabilities we have developed, we focus on Building Automation and Control networks (BACnet) from a private commercial building system. Our results show that “Model Zoo” built from binary classifiers based on each device or behavior combined with an ensemble classifier integrating information from all classifiers provides a reliable methodology to identify unknown devices as well as determining specific known devices when the device type is in the training set. The capability of the Model Zoo framework is shown to be directly linked to feature engineering and learning, and the dependency of the feature selection varies depending on both the binary and ensemble classifiers as well.

The ability to predict the onset of labour is seen to be an important tool in a clinical setting. Magnetomyography has shown promise in the area of labour imminency prediction, but its clinical application remains limited due to high resource consumption associated with its broad number of channels. In this study, five electrode channels, which account for 3.3% of the total, are used alongside a novel signal decomposition algorithm and low complexity classifiers (logistic regression and linear-SVM) to classify between labour imminency due within 0 to 48 hours and >48 hours. The results suggest that the parsimonious representation comprising of five electrode channels and novel signal decomposition method alongside the candidate classifiers could allow for greater affordability and hence clinical viability of the magnetomyography-based prediction model, which carries a good degree of model interpretability. The results showed around a 20% increase on average for the novel decomposition method, alongside a reduced group of features across the various classification metrics considered for both the logistic regression and support vector machine.