Understanding and explaining the predictions of Graph Neural Networks (GNNs), is crucial for enhancing their safety and trustworthiness. Subgraph-level explanations are gaining attention for their intuitive appeal. However, most existing subgraph-level explainers face efficiency challenges in explaining GNNs due to complex search processes. The key challenge is to find a balance between intuitiveness and efficiency while ensuring transparency. Additionally, these explainers usually induce subgraphs by nodes, which may introduce less-intuitive disconnected nodes in the subgraph-level explanations or omit many important subgraph structures. In this paper, we reveal that inducing subgraph explanations by edges is more comprehensive than other subgraph inducing techniques. We also emphasize the need of determining the subgraph explanation size for each data instance, as different data instances may involve different important substructures. Building upon these considerations, we introduce a training-free approach, named EiG-Search. We employ an efficient linear-time search algorithm over the edge-induced subgraphs, where the edges are ranked by an enhanced gradient-based importance. We conduct extensive experiments on a total of seven datasets, demonstrating its superior performance and efficiency both quantitatively and qualitatively over the leading baselines.
相關內容
Structure-Based Drug Design (SBDD) focuses on generating valid ligands that strongly and specifically bind to a designated protein pocket. Several methods use machine learning for SBDD to generate these ligands in 3D space, conditioned on the structure of a desired protein pocket. Recently, diffusion models have shown success here by modeling the underlying distributions of atomic positions and types. While these methods are effective in considering the structural details of the protein pocket, they often fail to explicitly consider the binding affinity. Binding affinity characterizes how tightly the ligand binds to the protein pocket, and is measured by the change in free energy associated with the binding process. It is one of the most crucial metrics for benchmarking the effectiveness of the interaction between a ligand and protein pocket. To address this, we propose BADGER: Binding Affinity Diffusion Guidance with Enhanced Refinement. BADGER is a general guidance method to steer the diffusion sampling process towards improved protein-ligand binding, allowing us to adjust the distribution of the binding affinity between ligands and proteins. Our method is enabled by using a neural network (NN) to model the energy function, which is commonly approximated by AutoDock Vina (ADV). ADV's energy function is non-differentiable, and estimates the affinity based on the interactions between a ligand and target protein receptor. By using a NN as a differentiable energy function proxy, we utilize the gradient of our learned energy function as a guidance method on top of any trained diffusion model. We show that our method improves the binding affinity of generated ligands to their protein receptors by up to 60\%, significantly surpassing previous machine learning methods. We also show that our guidance method is flexible and can be easily applied to other diffusion-based SBDD frameworks.
Early and accurate detection of anomalous events on the freeway, such as accidents, can improve emergency response and clearance. However, existing delays and errors in event identification and reporting make it a difficult problem to solve. Current large-scale freeway traffic datasets are not designed for anomaly detection and ignore these challenges. In this paper, we introduce the first large-scale lane-level freeway traffic dataset for anomaly detection. Our dataset consists of a month of weekday radar detection sensor data collected in 4 lanes along an 18-mile stretch of Interstate 24 heading toward Nashville, TN, comprising over 3.7 million sensor measurements. We also collect official crash reports from the Nashville Traffic Management Center and manually label all other potential anomalies in the dataset. To show the potential for our dataset to be used in future machine learning and traffic research, we benchmark numerous deep learning anomaly detection models on our dataset. We find that unsupervised graph neural network autoencoders are a promising solution for this problem and that ignoring spatial relationships leads to decreased performance. We demonstrate that our methods can reduce reporting delays by over 10 minutes on average while detecting 75% of crashes. Our dataset and all preprocessing code needed to get started are publicly released at //vu.edu/ft-aed/ to facilitate future research.
As automation in the field of automated driving (AD) progresses, ensuring the safety and functionality of AD functions (ADFs) becomes crucial. Virtual scenario-based testing has emerged as a prevalent method for evaluating these systems, allowing for a wider range of testing environments and reproducibility of results. This approach involves AD-equipped test vehicles operating within predefined scenarios to achieve specific driving objectives. To comprehensively assess the impact of road network properties on the performance of an ADF, varying parameters such as intersection angle, curvature and lane width is essential. However, covering all potential scenarios is impractical, necessitating the identification of feasible parameter ranges and automated generation of corresponding road networks for simulation. Automating the workflow of road network generation, parameter variation, simulation, and evaluation leads to a comprehensive understanding of an ADF's behavior in diverse road network conditions. This paper aims to investigate the influence of road network parameters on the performance of a prototypical ADF through virtual scenario-based testing, ultimately advocating the importance of road topology in assuring safety and reliability of ADFs.
Understanding the interdependence between autonomous and human-operated vehicles remains an ongoing challenge, with significant implications for the safety and feasibility of autonomous driving.This interdependence arises from inherent interactions among road users.Thus, it is crucial for Autonomous Vehicles (AVs) to understand and analyze the intentions of human-driven vehicles, and to display behavior comprehensible to other traffic participants.To this end, this paper presents GTP-UDRIVE, a unified game-theoretic trajectory planner and decision-maker considering a mixed-traffic environment. Our model considers the intentions of other vehicles in the decision-making process and provides the AV with a human-like trajectory, based on the clothoid interpolation technique.% This study investigates a solver based on Particle Swarm Optimization (PSO) that quickly converges to an optimal decision.Among highly interactive traffic scenarios, the intersection crossing is particularly challenging. Hence, we choose to demonstrate the feasibility and effectiveness of our method in real traffic conditions, using an experimental autonomous vehicle at an unsignalized intersection. Testing results reveal that our approach is suitable for 1) Making decisions and generating trajectories simultaneously. 2) Describing the vehicle's trajectory as a piecewise clothoid and enforcing geometric constraints. 3) Reducing search space dimensionality for the trajectory optimization problem.
The security concerns surrounding Large Language Models (LLMs) have been extensively explored, yet the safety of Multimodal Large Language Models (MLLMs) remains understudied. In this paper, we observe that Multimodal Large Language Models (MLLMs) can be easily compromised by query-relevant images, as if the text query itself were malicious. To address this, we introduce MM-SafetyBench, a comprehensive framework designed for conducting safety-critical evaluations of MLLMs against such image-based manipulations. We have compiled a dataset comprising 13 scenarios, resulting in a total of 5,040 text-image pairs. Our analysis across 12 state-of-the-art models reveals that MLLMs are susceptible to breaches instigated by our approach, even when the equipped LLMs have been safety-aligned. In response, we propose a straightforward yet effective prompting strategy to enhance the resilience of MLLMs against these types of attacks. Our work underscores the need for a concerted effort to strengthen and enhance the safety measures of open-source MLLMs against potential malicious exploits. The resource is available at //github.com/isXinLiu/MM-SafetyBench
Magnetic Particle Imaging is an emerging imaging modality through which it is possible to detect tracers containing superparamagnetic nanoparticles. The exposure of the particles to dynamic magnetic fields generates a non-linear response that is used to locate the particles and produce an image of their distribution. The bounding box that can be covered by a single scan curve depends on the strength of the gradients of the magnetic fields applied, which is limited due to the risk of causing peripheral nerve stimulation (PNS) in the patients. To address this issue, multiple scans are performed. The scan data must be merged together to produce reconstructions of larger regions of interest. In this paper we propose a mathematical framework which can deal with rather general multi-patching scenarios including rigid transformations of the field of view (FoV), the specimen and of the scanner. We show the flexibility of this framework in a variety of different scanning scenarios. Moreover, we describe an iterative reconstruction algorithm that yields a reconstruction of the target distribution by minimizing a convex functional which includes positivity constraints and sparsity enforcing priors. We show its convergence to a minimizer and perform numerical experiments on simulated data.
Reward Machines provide an automata-inspired structure for specifying instructions, safety constraints, and other temporally extended reward-worthy behaviour. By exposing complex reward function structure, they enable counterfactual learning updates that have resulted in impressive sample efficiency gains. While Reward Machines have been employed in both tabular and deep RL settings, they have typically relied on a ground-truth interpretation of the domain-specific vocabulary that form the building blocks of the reward function. Such ground-truth interpretations can be elusive in many real-world settings, due in part to partial observability or noisy sensing. In this paper, we explore the use of Reward Machines for Deep RL in noisy and uncertain environments. We characterize this problem as a POMDP and propose a suite of RL algorithms that leverage task structure under uncertain interpretation of domain-specific vocabulary. Theoretical analysis exposes pitfalls in naive approaches to this problem, while experimental results show that our algorithms successfully leverage task structure to improve performance under noisy interpretations of the vocabulary. Our results provide a general framework for exploiting Reward Machines in partially observable environments.
We introduce TorchOpera, a compound AI system for enhancing the safety and quality of prompts and responses for Large Language Models. TorchOpera ensures that all user prompts are safe, contextually grounded, and effectively processed, while enhancing LLM responses to be relevant and high quality. TorchOpera utilizes the vector database for contextual grounding, rule-based wrappers for flexible modifications, and specialized mechanisms for detecting and adjusting unsafe or incorrect content. We also provide a view of the compound AI system to reduce the computational cost. Extensive experiments show that TorchOpera ensures the safety, reliability, and applicability of LLMs in real-world settings while maintaining the efficiency of LLM responses.
Despite significant progress in autonomous vehicles (AVs), the development of driving policies that ensure both the safety of AVs and traffic flow efficiency has not yet been fully explored. In this paper, we propose an enhanced human-in-the-loop reinforcement learning method, termed the Human as AI mentor-based deep reinforcement learning (HAIM-DRL) framework, which facilitates safe and efficient autonomous driving in mixed traffic platoon. Drawing inspiration from the human learning process, we first introduce an innovative learning paradigm that effectively injects human intelligence into AI, termed Human as AI mentor (HAIM). In this paradigm, the human expert serves as a mentor to the AI agent. While allowing the agent to sufficiently explore uncertain environments, the human expert can take control in dangerous situations and demonstrate correct actions to avoid potential accidents. On the other hand, the agent could be guided to minimize traffic flow disturbance, thereby optimizing traffic flow efficiency. In detail, HAIM-DRL leverages data collected from free exploration and partial human demonstrations as its two training sources. Remarkably, we circumvent the intricate process of manually designing reward functions; instead, we directly derive proxy state-action values from partial human demonstrations to guide the agents' policy learning. Additionally, we employ a minimal intervention technique to reduce the human mentor's cognitive load. Comparative results show that HAIM-DRL outperforms traditional methods in driving safety, sampling efficiency, mitigation of traffic flow disturbance, and generalizability to unseen traffic scenarios. The code and demo videos for this paper can be accessed at: //zilin-huang.github.io/HAIM-DRL-website/
Trustworthy capability evaluations are crucial for ensuring the safety of AI systems, and are becoming a key component of AI regulation. However, the developers of an AI system, or the AI system itself, may have incentives for evaluations to understate the AI's actual capability. These conflicting interests lead to the problem of sandbagging $\unicode{x2013}$ which we define as "strategic underperformance on an evaluation". In this paper we assess sandbagging capabilities in contemporary language models (LMs). We prompt frontier LMs, like GPT-4 and Claude 3 Opus, to selectively underperform on dangerous capability evaluations, while maintaining performance on general (harmless) capability evaluations. Moreover, we find that models can be fine-tuned, on a synthetic dataset, to hide specific capabilities unless given a password. This behaviour generalizes to high-quality, held-out benchmarks such as WMDP. In addition, we show that both frontier and smaller models can be prompted, or password-locked, to target specific scores on a capability evaluation. Even more, we found that a capable password-locked model (Llama 3 70b) is reasonably able to emulate a less capable model (Llama 2 7b). Overall, our results suggest that capability evaluations are vulnerable to sandbagging. This vulnerability decreases the trustworthiness of evaluations, and thereby undermines important safety decisions regarding the development and deployment of advanced AI systems.
小貼士
登錄享
相關主題
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191