Artificial Intelligence (cs.AI)

Abstract: The spread of infectious diseases, rumors, and harmful speech in networks can result in substantial losses, underscoring the significance of studying how to suppress such hazardous events. However, previous studies often assume full knowledge of the network structure, which is often not the case in real-world scenarios. In this paper, we address the challenge of controlling the propagation of hazardous events by removing nodes when the network structure is unknown. To tackle this problem, we propose a hierarchical reinforcement learning method that drastically reduces the action space, making the problem feasible to solve. Simulation experiments demonstrate the superiority of our method over the baseline methods. Remarkably, even though the baseline methods possess extensive knowledge of the network structure, while our method has no prior information about it, our approach still achieves better results.

Abstract: Data Spaces are an emerging concept for the trusted implementation of data-based applications and business models, offering a high degree of flexibility and sovereignty to all stakeholders. As Data Spaces are currently emerging in different domains such as mobility, health or food, semantic interfaces need to be identified and implemented to ensure the technical interoperability of these Data Spaces. This paper consolidates data models from 13 different domains and analyzes the ontological dissonance of these domains. Using a network graph, central data models and ontology attributes are identified, while the semantic heterogeneity of these domains is described qualitatively. The research outlook describes how these results help to connect different Data Spaces across domains.

Abstract: Large Language Models (LLMs) such as ChatGPT have received enormous attention over the past year and are now used by hundreds of millions of people every day. The rapid adoption of this technology naturally raises questions about the possible biases such models might exhibit. In this work, we tested one of these models (GPT-3) on a range of cognitive effects, which are systematic patterns that are usually found in human cognitive tasks. We found that LLMs are indeed prone to several human cognitive effects. Specifically, we show that the priming, distance, SNARC, and size congruity effects were presented with GPT-3, while the anchoring effect is absent. We describe our methodology, and specifically the way we converted real-world experiments to text-based experiments. Finally, we speculate on the possible reasons why GPT-3 exhibits these effects and discuss whether they are imitated or reinvented.

Abstract: Human emotion understanding is pivotal in making conversational technology mainstream. We view speech emotion understanding as a perception task which is a more realistic setting. With varying contexts (languages, demographics, etc.) different share of people perceive the same speech segment as a non-unanimous emotion. As part of the ACM Multimedia 2023 Computational Paralinguistics ChallengE (ComParE) in the EMotion Share track, we leverage their rich dataset of multilingual speakers and multi-label regression target of 'emotion share' or perception of that emotion. We demonstrate that the training scheme of different foundation models dictates their effectiveness for tasks beyond speech recognition, especially for non-semantic speech tasks like emotion understanding. This is a very complex task due to multilingual speakers, variability in the target labels, and inherent imbalance in the regression dataset. Our results show that HuBERT-Large with a self-attention-based light-weight sequence model provides 4.6% improvement over the reported baseline.

Abstract: In today's landscape, smartphones have evolved into hubs for hosting a multitude of deep learning models aimed at local execution. A key realization driving this work is the notable fragmentation among these models, characterized by varied architectures, operators, and implementations. This fragmentation imposes a significant burden on the comprehensive optimization of hardware, system settings, and algorithms. Buoyed by the recent strides in large foundation models, this work introduces a pioneering paradigm for mobile AI: a collaborative management approach between the mobile OS and hardware, overseeing a foundational model capable of serving a broad spectrum of mobile AI tasks, if not all. This foundational model resides within the NPU and remains impervious to app or OS revisions, akin to firmware. Concurrently, each app contributes a concise, offline fine-tuned "adapter" tailored to distinct downstream tasks. From this concept emerges a concrete instantiation known as \sys. It amalgamates a curated selection of publicly available Large Language Models (LLMs) and facilitates dynamic data flow. This concept's viability is substantiated through the creation of an exhaustive benchmark encompassing 38 mobile AI tasks spanning 50 datasets, including domains such as Computer Vision (CV), Natural Language Processing (NLP), audio, sensing, and multimodal inputs. Spanning this benchmark, \sys unveils its impressive performance. It attains accuracy parity in 85\% of tasks, demonstrates improved scalability in terms of storage and memory, and offers satisfactory inference speed on Commercial Off-The-Shelf (COTS) mobile devices fortified with NPU support. This stands in stark contrast to task-specific models tailored for individual applications.

Abstract: Modern communication systems rely on accurate channel estimation to achieve efficient and reliable transmission of information. As the communication channel response is highly related to the user's location, one can use a neural network to map the user's spatial coordinates to the channel coefficients. However, these latter are rapidly varying as a function of the location, on the order of the wavelength. Classical neural architectures being biased towards learning low frequency functions (spectral bias), such mapping is therefore notably difficult to learn. In order to overcome this limitation, this paper presents a frugal, model-based network that separates the low frequency from the high frequency components of the target mapping function. This yields an hypernetwork architecture where the neural network only learns low frequency sparse coefficients in a dictionary of high frequency components. Simulation results show that the proposed neural network outperforms standard approaches on realistic synthetic data.

Abstract: The latest cancer statistics indicate a decrease in cancer-related mortality. However, due to the growing and ageing population, the absolute number of people living with cancer is set to keep increasing. This paper presents ASCAPE, an open AI infrastructure that takes advantage of the recent advances in Artificial Intelligence (AI) and Machine Learning (ML) to support cancer patients quality of life (QoL). With ASCAPE health stakeholders (e.g. hospitals) can locally process their private medical data and then share the produced knowledge (ML models) through the open AI infrastructure.

Abstract: In current ML field models are getting larger and more complex, data we use are also getting larger in quantity and higher in dimension, so in order to train better models, save training time and computational resources, a good Feature Selection (FS) method in preprocessing stage is necessary. Feature importance (FI) is of great importance since it is the basis of feature selection. This paper creatively introduces the calculation of PNS(the probability of Necessity and Sufficiency) in Causality into quantifying feature importance and creates new FI measuring methods: PN-FI, which means how much importance a feature has in image recognition tasks, PS_FI that means how much importance a feature has in image generating tasks, and PNS_FI which measures both. The main body of this paper is three RCTs, with whose results we show how PS_FI, PN_FI and PNS_FI of three features: dog nose, dog eyes and dog mouth are calculated. The FI values are intervals with tight upper and lower bounds.

Abstract: Process pattern discovery methods (PPDMs) aim at identifying patterns of interest to users. Existing PPDMs typically are unsupervised and focus on a single dimension of interest, such as discovering frequent patterns. We present an interactive multi interest driven framework for process pattern discovery aimed at identifying patterns that are optimal according to a multi-dimensional analysis goal. The proposed approach is iterative and interactive, thus taking experts knowledge into account during the discovery process. The paper focuses on a concrete analysis goal, i.e., deriving process patterns that affect the process outcome. We evaluate the approach on real world event logs in both interactive and fully automated settings. The approach extracted meaningful patterns validated by expert knowledge in the interactive setting. Patterns extracted in the automated settings consistently led to prediction performance comparable to or better than patterns derived considering single interest dimensions without requiring user defined thresholds.

Abstract: Computational agents support humans in many areas of life and are therefore found in heterogeneous contexts. This means that agents operate in rapidly changing environments and can be confronted with huge state and action spaces. In order to perform services and carry out activities in a goal-oriented manner, agents require prior knowledge and therefore have to develop and pursue context-dependent policies. The problem is that prescribing policies in advance is limited and inflexible, especially in dynamically changing environments. Moreover, the context of an agent determines its choice of actions. Since the environments in which agents operate can be stochastic and complex in terms of the number of states and feasible actions, activities are usually modelled in a simplified way by Markov decision processes so that agents with reinforcement learning are able to learn policies that help to capture the context and act accordingly to optimally perform activities. However, training policies for all possible contexts using reinforcement learning is time-consuming. A requirement and challenge for agents is to learn strategies quickly and respond immediately in cross-context environments and applications. In this work, we propose a novel simulation-based approach that enables a) the representation of heterogeneous contexts through knowledge graphs and entity embeddings and b) the context-aware composition of policies on demand by ensembles of agents running in parallel. The evaluation we performed on the "Virtual Home" dataset indicates that agents that need to seamlessly switch between different contexts, can request on-the-fly composed policies that lead to the successful completion of context-appropriate activities without having to learn these policies in lengthy training steps and episodes, in contrast to agents that apply reinforcement learning.

Abstract: Autonomous vehicles are advanced driving systems that are well known for being vulnerable to various adversarial attacks, compromising the vehicle's safety, and posing danger to other road users. Rather than actively training complex adversaries by interacting with the environment, there is a need to first intelligently find and reduce the search space to only those states where autonomous vehicles are found less confident. In this paper, we propose a blackbox testing framework ReMAV using offline trajectories first to analyze the existing behavior of autonomous vehicles and determine appropriate thresholds for finding the probability of failure events. Our reward modeling technique helps in creating a behavior representation that allows us to highlight regions of likely uncertain behavior even when the baseline autonomous vehicle is performing well. This approach allows for more efficient testing without the need for computational and inefficient active adversarial learning techniques. We perform our experiments in a high-fidelity urban driving environment using three different driving scenarios containing single and multi-agent interactions. Our experiment shows 35%, 23%, 48%, and 50% increase in occurrences of vehicle collision, road objects collision, pedestrian collision, and offroad steering events respectively by the autonomous vehicle under test, demonstrating a significant increase in failure events. We also perform a comparative analysis with prior testing frameworks and show that they underperform in terms of training-testing efficiency, finding total infractions, and simulation steps to identify the first failure compared to our approach. The results show that the proposed framework can be used to understand existing weaknesses of the autonomous vehicles under test in order to only attack those regions, starting with the simplistic perturbation models.

Abstract: As technology progresses, industrial and scientific robots are increasingly being used in diverse settings. In many cases, however, programming the robot to perform such tasks is technically complex and costly. To maximize the utility of robots in industrial and scientific settings, they require the ability to quickly shift from one task to another. Reinforcement learning algorithms provide the potential to enable robots to learn optimal solutions to complete new tasks without directly reprogramming them. The current state-of-the-art in reinforcement learning, however, generally relies on fast simulations and parallelization to achieve optimal performance. These are often not possible in robotics applications. Thus, a significant amount of research is required to facilitate the efficient and safe, training and deployment of industrial and scientific reinforcement learning robots. This technical report outlines our initial research into the application of deep reinforcement learning on an industrial UR10e robot. The report describes the reinforcement learning environments created to facilitate policy learning with the UR10e, a robotic arm from Universal Robots, and presents our initial results in training deep Q-learning and proximal policy optimization agents on the developed reinforcement learning environments. Our results show that proximal policy optimization learns a better, more stable policy with less data than deep Q-learning. The corresponding code for this work is available at \url{https://github.com/cbellinger27/bendRL_reacher_tracker}

Abstract: Childhood and adolescent obesity rates are a global concern because obesity is associated with chronic diseases and long-term health risks. Artificial intelligence technology has emerged as a promising solution to accurately predict obesity rates and provide personalized feedback to adolescents. This study emphasizes the importance of early identification and prevention of obesity-related health issues. Factors such as height, weight, waist circumference, calorie intake, physical activity levels, and other relevant health information need to be considered for developing robust algorithms for obesity rate prediction and delivering personalized feedback. Hence, by collecting health datasets from 321 adolescents, we proposed an adolescent obesity prediction system that provides personalized predictions and assists individuals in making informed health decisions. Our proposed deep learning framework, DeepHealthNet, effectively trains the model using data augmentation techniques, even when daily health data are limited, resulting in improved prediction accuracy (acc: 0.8842). Additionally, the study revealed variations in the prediction of the obesity rate between boys (acc: 0.9320) and girls (acc: 0.9163), allowing the identification of disparities and the determination of the optimal time to provide feedback. The proposed system shows significant potential in effectively addressing childhood and adolescent obesity.

Abstract: We encounter time series data in many domains such as finance, physics, business, and weather. One of the main tasks of time series analysis, one that helps to take informed decisions under uncertainty, is forecasting. Time series are often hierarchically structured, e.g., a company sales might be broken down into different regions, and each region into different stores. In some cases the number of series in the hierarchy is too big to fit in a single model to produce forecasts in relevant time, and a decentralized approach is beneficial. One way to do this is to train independent forecasting models for each series and for some summary statistics series implied by the hierarchy (e.g. the sum of all series) and to pass those models to a reconciliation algorithm to improve those forecasts by sharing information between the series. In this work we focus on the reconciliation step, and propose a method to do so from a Bayesian perspective - Bayesian forecast reconciliation. We also define the common case of linear Gaussian reconciliation, where the forecasts are Gaussian and the hierarchy has linear structure, and show that we can compute reconciliation in closed form. We evaluate these methods on synthetic and real data sets, and compare them to other work in this field.

Abstract: This paper aims to investigate the mathematical problem-solving capabilities of Chat Generative Pre-Trained Transformer (ChatGPT) in case of Bayesian reasoning. The study draws inspiration from Zhu & Gigerenzer's research in 2006, which posed the question: Can children reason the Bayesian way? In the pursuit of answering this question, a set of 10 Bayesian reasoning problems were presented. The results of their work revealed that children's ability to reason effectively using Bayesian principles is contingent upon a well-structured information representation. In this paper, we present the same set of 10 Bayesian reasoning problems to ChatGPT. Remarkably, the results demonstrate that ChatGPT provides the right solutions to all problems.