Search

This report presents a comprehensive response to the United Nation's Interim Report on Governing Artificial Intelligence (AI) for Humanity. It emphasizes the transformative potential of AI in achieving the Sustainable Development Goals (SDGs) while acknowledging the need for robust governance to mitigate associated risks. The response highlights opportunities for promoting equitable, secure, and inclusive AI ecosystems, which should be supported by investments in infrastructure and multi-stakeholder collaborations across jurisdictions. It also underscores challenges, including societal inequalities exacerbated by AI, ethical concerns, and environmental impacts. Recommendations advocate for legally binding norms, transparency, and multi-layered data governance models, alongside fostering AI literacy and capacity-building initiatives. Internationally, the report calls for harmonising AI governance frameworks with established laws, human rights standards, and regulatory approaches. The report concludes with actionable principles for fostering responsible AI governance through collaboration among governments, industry, academia, and civil society, ensuring the development of AI aligns with universal human values and the public good., Comment: Submitted to United Nations. 23 pages. All the Authors Contributed Equally

Modern software-based systems operate under rapidly changing conditions and face ever-increasing uncertainty. In response, systems are increasingly adaptive and reliant on artificial-intelligence methods. In addition to the ubiquity of software with respect to users and application areas (e.g., transportation, smart grids, medicine, etc.), these high-impact software systems necessarily draw from many disciplines for foundational principles, domain expertise, and workflows. Recent progress with lowering the barrier to entry for coding has led to a broader community of developers, who are not necessarily software engineers. As such, the field of software engineering needs to adapt accordingly and offer new methods to systematically develop high-quality software systems by a broad range of experts and non-experts. This paper looks at these new challenges and proposes to address them through the lens of Abstraction. Abstraction is already used across many disciplines involved in software development -- from the time-honored classical deductive reasoning and formal modeling to the inductive reasoning employed by modern data science. The software engineering of the future requires Abstraction Engineering -- a systematic approach to abstraction across the inductive and deductive spaces. We discuss the foundations of Abstraction Engineering, identify key challenges, highlight the research questions that help address these challenges, and create a roadmap for future research.

We present foundation language models developed to power Apple Intelligence features, including a ~3 billion parameter model designed to run efficiently on devices and a large server-based language model designed for Private Cloud Compute. These models are designed to perform a wide range of tasks efficiently, accurately, and responsibly. This report describes the model architecture, the data used to train the model, the training process, how the models are optimized for inference, and the evaluation results. We highlight our focus on Responsible AI and how the principles are applied throughout the model development.

Humans rely on strong inductive biases to learn from few examples and abstract useful information from sensory data. Instilling such biases in machine learning models has been shown to improve their performance on various benchmarks including few-shot learning, robustness, and alignment. However, finding effective training procedures to achieve that goal can be challenging as psychologically-rich training data such as human similarity judgments are expensive to scale, and Bayesian models of human inductive biases are often intractable for complex, realistic domains. Here, we address this challenge by introducing a Bayesian notion of generative similarity whereby two datapoints are considered similar if they are likely to have been sampled from the same distribution. This measure can be applied to complex generative processes, including probabilistic programs. We show that generative similarity can be used to define a contrastive learning objective even when its exact form is intractable, enabling learning of spatial embeddings that express specific inductive biases. We demonstrate the utility of our approach by showing that it can be used to capture human inductive biases for geometric shapes, distinguish different abstract drawing styles that are parameterized by probabilistic programs, and capture abstract high-level categories that enable generalization.

Most applications of machine learning to classification assume a closed set of balanced classes. This is at odds with the real world, where class occurrence statistics often follow a long-tailed power-law distribution and it is unlikely that all classes are seen in a single sample. Nonparametric Bayesian models naturally capture this phenomenon, but have significant practical barriers to widespread adoption, namely implementation complexity and computational inefficiency. To address this, we present a method for extracting the inductive bias from a nonparametric Bayesian model and transferring it to an artificial neural network. By simulating data with a nonparametric Bayesian prior, we can metalearn a sequence model that performs inference over an unlimited set of classes. After training, this "neural circuit" has distilled the corresponding inductive bias and can successfully perform sequential inference over an open set of classes. Our experimental results show that the metalearned neural circuit achieves comparable or better performance than particle filter-based methods for inference in these models while being faster and simpler to use than methods that explicitly incorporate Bayesian nonparametric inference., Comment: 13 pages, 3 figures. Code available at https://github.com/jakesnell/neural-circuits

Bayesian filtering approximates the true underlying behavior of a time-varying system by inverting an explicit generative model to convert noisy measurements into state estimates. This process typically requires either storage, inversion, and multiplication of large matrices or Monte Carlo estimation, neither of which are practical in high-dimensional state spaces such as the weight spaces of artificial neural networks. Here, we frame the standard Bayesian filtering problem as optimization over a time-varying objective. Instead of maintaining matrices for the filtering equations or simulating particles, we specify an optimizer that defines the Bayesian filter implicitly. In the linear-Gaussian setting, we show that every Kalman filter has an equivalent formulation using K steps of gradient descent. In the nonlinear setting, our experiments demonstrate that our framework results in filters that are effective, robust, and scalable to high-dimensional systems, comparing well against the standard toolbox of Bayesian filtering solutions. We suggest that it is easier to fine-tune an optimizer than it is to specify the correct filtering equations, making our framework an attractive option for high-dimensional filtering problems., Comment: Under review at ICLR 2024

In an ever-changing world, even software that fulfils its requirements may have un-envisioned aftereffects with significant impacts. We explored how such impacts can be better understood at the pre-design phase in support of organisational preparedness. We considered three real-world case studies and engaged with literature from several disciplines to develop a conceptual framework. Across three workshops with industry practitioners and academics creative strategies from speculative design practices were used to prompt engagement with the framework. We found participant groups navigated the model with either a convergent or divergent intent. The academics, operating in an exploratory mode, came to a broad understanding of a class of technologies through its impacts. Operating in an anticipatory mode the industry practitioners came to a specific understanding of a technology's potential in their workplace. The study demonstrated potential for the conceptual framework to be used as a tool with implications for research and practice., Comment: 14 pages, 2 figures

A Reinforcement Learning (RL) system depends on a set of initial conditions (hyperparameters) that affect the system's performance. However, defining a good choice of hyperparameters is a challenging problem. Hyperparameter tuning often requires manual or automated searches to find optimal values. Nonetheless, a noticeable limitation is the high cost of algorithm evaluation for complex models, making the tuning process computationally expensive and time-consuming. In this paper, we propose a framework based on integrating complex event processing and temporal models, to alleviate these trade-offs. Through this combination, it is possible to gain insights about a running RL system efficiently and unobtrusively based on data stream monitoring and to create abstract representations that allow reasoning about the historical behaviour of the RL system. The obtained knowledge is exploited to provide feedback to the RL system for optimising its hyperparameters while making effective use of parallel resources. We introduce a novel history-aware epsilon-greedy logic for hyperparameter optimisation that instead of using static hyperparameters that are kept fixed for the whole training, adjusts the hyperparameters at runtime based on the analysis of the agent's performance over time windows in a single agent's lifetime. We tested the proposed approach in a 5G mobile communications case study that uses DQN, a variant of RL, for its decision-making. Our experiments demonstrated the effects of hyperparameter tuning using history on training stability and reward values. The encouraging results show that the proposed history-aware framework significantly improved performance compared to traditional hyperparameter tuning approaches.

How might we design Reinforcement Learning (RL)-based recommenders that encourage aligning user trajectories with the underlying user satisfaction? Three research questions are key: (1) measuring user satisfaction, (2) combatting sparsity of satisfaction signals, and (3) adapting the training of the recommender agent to maximize satisfaction. For measurement, it has been found that surveys explicitly asking users to rate their experience with consumed items can provide valuable orthogonal information to the engagement/interaction data, acting as a proxy to the underlying user satisfaction. For sparsity, i.e, only being able to observe how satisfied users are with a tiny fraction of user-item interactions, imputation models can be useful in predicting satisfaction level for all items users have consumed. For learning satisfying recommender policies, we postulate that reward shaping in RL recommender agents is powerful for driving satisfying user experiences. Putting everything together, we propose to jointly learn a policy network and a satisfaction imputation network: The role of the imputation network is to learn which actions are satisfying to the user; while the policy network, built on top of REINFORCE, decides which items to recommend, with the reward utilizing the imputed satisfaction. We use both offline analysis and live experiments in an industrial large-scale recommendation platform to demonstrate the promise of our approach for satisfying user experiences., Comment: Accepted in Reinforcement Learning for Real Life (RL4RealLife) Workshop in the 38th International Conference on Machine Learning, 2021

Relaxation Runge-Kutta methods reproduce a fully discrete dissipation (or conservation) of entropy for entropy stable semi-discretizations of nonlinear conservation laws. In this paper, we derive the discrete adjoint of relaxation Runge-Kutta schemes, which are applicable to discretize-then-optimize approaches for optimal control problems. Furthermore, we prove that the derived discrete relaxation Runge-Kutta adjoint preserves time-symmetry when applied to linear skew-symmetric systems of ODEs. Numerical experiments verify these theoretical results while demonstrating the importance of appropriately treating the relaxation parameter when computing the discrete adjoint., Comment: 39 pages, 7 figures (not including subfigures). Submitted to the Journal of Computational Physics

Decision-making for self-adaptation approaches need to address different challenges, including the quantification of the uncertainty of events that cannot be foreseen in advance and their effects, and dealing with conflicting objectives that inherently involve multi-objective decision making (e.g., avoiding costs vs. providing reliable service). To enable researchers to evaluate and compare decision-making techniques for self-adaptation, we present the RDMSim exemplar. RDMSim enables researchers to evaluate and compare techniques for decision-making under environmental uncertainty that support self-adaptation. The focus of the exemplar is on the domain problem related to Remote Data Mirroring, which gives opportunity to face the challenges described above. RDMSim provides probe and effector components for easy integration with external adaptation managers, which are associated with decision-making techniques and based on the MAPE-K loop. Specifically, the paper presents (i) RDMSim, a simulator for real-world experimentation, (ii) a set of realistic simulation scenarios that can be used for experimentation and comparison purposes, (iii) data for the sake of comparison., Comment: Accepted at the 16th International Symposium on Software Engineering for Adaptive and Self-Managing Systems (SEAMS 2021)

The mobility of people is at the center of transportation planning and decision-making of the cities of the future. In order to accelerate the transition to zero-emissions and to maximize air quality benefits, smart cities are prioritizing walking, cycling, shared mobility services and public transport over the use of private cars. Extensive progress has been made in autonomous and electric cars. Autonomous Vehicles (AV) are increasingly capable of moving without full control of humans, automating some aspects of driving, such as steering or braking. For these reasons, cities are investing in the infrastructure and technology needed to support connected, multi-modal transit networks that include shared electric Autonomous Vehicles (AV). The relationship between traditional public transport and new mobility services is in the spotlight and need to be rethought. This paper proposes an agent-based simulation framework that allows for the creation and simulation of mobility scenarios to investigate the impact of new mobility modes on a city daily life. It lets traffic planners explore the cooperative integration of AV using a decentralized control approach. A prototype has been implemented and validated with data of the city of Trento., Comment: 13 pages, 10 Figures, Early Access Article published at the IEEE Transactions on Intelligent Transportation Systems

Motivational digital systems offer capabilities to engage and motivate end-users to foster behavioral changes towards a common goal. In general these systems use gamification principles in non-games contexts. Over the years, gamification has gained consensus among researchers and practitioners as a tool to motivate people to perform activities with the ultimate goal of promoting behavioural change, or engaging the users to perform activities that can offer relevant benefits but which can be seen as unrewarding and even tedious. There exists a plethora of heterogeneous application scenarios towards reaching the common good that can benefit from gamification. However, an open problem is how to effectively combine multiple motivational campaigns to maximise the degree of participation without exposing the system to counterproductive behaviours. We conceive motivational digital systems as multi-agent systems: self-adaptation is a feature of the overall system, while individual agents may self-adapt in order to leverage other agents' resources, functionalities and capabilities to perform tasks more efficiently and effectively. Consequently, multiple campaigns can be run and adapted to reach common good. At the same time, agents are grouped into micro-communities in which agents contribute with their own social capital and leverage others' capabilities to balance their weaknesses. In this paper we propose our vision on how the principles at the base of the autonomous and multi-agent systems can be exploited to design multi-challenge motivational systems to engage smart communities towards common goals. We present an initial version of a general framework based on the MAPE-K loop and a set of research challenges that characterise our research roadmap for the implementation of our vision., Comment: 7 pages, 1 figure

Reinforcement learning algorithms in multi-agent systems deliver highly resilient and adaptable solutions for common problems in telecommunications,aerospace, and industrial robotics. However, achieving an optimal global goal remains a persistent obstacle for collaborative multi-agent systems, where learning affects the behaviour of more than one agent. A number of nonlinear function approximation methods have been proposed for solving the Bellman equation, which describe a recursive format of an optimal policy. However, how to leverage the value distribution based on reinforcement learning, and how to improve the efficiency and efficacy of such systems remain a challenge. In this work, we developed a reward-reinforced generative adversarial network to represent the distribution of the value function, replacing the approximation of Bellman updates. We demonstrated our method is resilient and outperforms other conventional reinforcement learning methods. This method is also applied to a practical case study: maximising the number of user connections to autonomous airborne base stations in a mobile communication network. Our method maximises the data likelihood using a cost function under which agents have optimal learned behaviours. This reward-reinforced generative adversarial network can be used as ageneric framework for multi-agent learning at the system level, Comment: 9 pages, 9 figures

This paper describes an investigation into value-based software engineering and proposes a comprehensive value taxonomy with an interpretation of design feature implications. The value taxonomy is used to assess the design of Covid19 symptom tracker applications., Comment: ICSE SEIS 2021

High-order entropy-stable discontinuous Galerkin (DG) methods for nonlinear conservation laws reproduce a discrete entropy inequality by combining entropy conservative finite volume fluxes with summation-by-parts (SBP) discretization matrices. In the DG context, on tensor product (quadrilateral and hexahedral) elements, SBP matrices are typically constructed by collocating at Lobatto quadrature points. Recent work has extended the construction of entropy-stable DG schemes to collocation at more accurate Gauss quadrature points. In this work, we extend entropy-stable Gauss collocation schemes to non-conforming meshes. Entropy-stable DG schemes require computing entropy conservative numerical fluxes between volume and surface quadrature nodes. On conforming tensor product meshes where volume and surface nodes are aligned, flux evaluations are required only between "lines" of nodes. However, on non-conforming meshes, volume and surface nodes are no longer aligned, resulting in a larger number of flux evaluations. We reduce this expense by introducing an entropy-stable mortar-based treatment of non-conforming interfaces via a face-local correction term, and provide necessary conditions for high-order accuracy. Numerical experiments in both two and three dimensions confirm the stability and accuracy of this approach., Comment: Submitted to Journal of Scientific Computing

Providing assurances for self-adaptive systems is challenging. A primary underlying problem is uncertainty that may stem from a variety of different sources, ranging from incomplete knowledge to sensor noise and uncertain behavior of humans in the loop. Providing assurances that the self-adaptive system complies with its requirements calls for an enduring process spanning the whole lifetime of the system. In this process, humans and the system jointly derive and integrate new evidence and arguments, which we coined perpetual assurances for self-adaptive systems. In this paper, we provide a background framework and the foundation for perpetual assurances for self-adaptive systems. We elaborate on the concrete challenges of offering perpetual assurances, requirements for solutions, realization techniques and mechanisms to make solutions suitable. We also present benchmark criteria to compare solutions. We then present a concrete exemplar that researchers can use to assess and compare approaches for perpetual assurances for self-adaptation.

We have measured the half-life of 30S, the parent of a superallowed 0+-to-0+ beta transition, to high precision using very pure sources and a 4pi proportional gas counter to detect the decay positrons. Our result for the half-life is 1.17992(34) s. As a byproduct of this measurement, we determined the half-life of its daughter, 30P, to be 2.501(2) min., Comment: 9 pages, 7 figures

We present the full description of a measurement of the branching ratios for the beta-decay of 38Ca. This decay includes five allowed 0+ --> 1+ branches and a superallowed 0+ --> 0+ one. With our new result for the latter, we determine its ft value to be 3062.3(68) s, a result whose precision (0.2%) is comparable to the precision of the thirteen well known 0+ --> 0+ transitions used up till now for the determination of Vud, the up-down quark-mixing element of the CKM matrix. The 38Ca superallowed transition thus becomes the first addition to this set of transitions in nearly a decade and the first for which a precise mirror comparison is possible, thus enabling an improved test of the isospin-symmetry-breaking corrections required for the extraction of Vud., Comment: 14 pages, 8 figures

We report the first branching-ratio measurement of the superallowed 0+-to-0+ beta transition from 38Ca. The result, 0.7728(16), leads to an ft value of 3062.3(68)s with a relative precision of +/-0.2%. This makes possible a high-precision comparison of the ft values for the mirror superallowed transitions, 38Ca-to-38mK and 38mK-to-38Ar, which sensitively tests the isospin symmetry-breaking corrections required to extract Vud, the up-down quark-mixing element of the Cabibbo-Kobayashi-Maskawa (CKM) matrix, from superallowed beta decay. The result supports the corrections currently used, and points the way to even tighter constraints on CKM unitarity., Comment: 5 pages, 3 figures

We have measured the K-shell internal conversion coefficient, alpha-K, for the 65.7-keV M4 transition in 119Sn to be 1621(25). This result agrees well with Dirac-Fock calculations in which the effect of the K-shell atomic vacancy is accounted for, and disagrees with calculations in which the vacancy is ignored. This extends our precision tests of theory to Z = 50, the lowest Z yet measured., Comment: 10 pages, 7 figures

The paper presents an algorithm for control and safe landing of impaired multi-rotor drones when one or more motors fail simultaneously or in any sequence. It includes three main components: an identification block, a reconfigurable control block, and a decisions making block. The identification block monitors each motor load characteristics and the current drawn, based on which the failures are detected. The control block generates the required total thrust and three axis torques for the altitude, horizontal position and/or orientation control of the drone based on the time scale separation and nonlinear dynamic inversion. The horizontal displacement is controlled by modulating the roll and pitch angles. The decision making algorithm maps the total thrust and three torques into the individual motor thrusts based on the information provided by the identification block. The drone continues the mission execution as long as the number of functioning motors provide controllability of it. Otherwise, the controller is switched to the safe mode, which gives up the yaw control, commands a safe landing spot and descent rate while maintaining the horizontal attitude.