This paper offers a roadmap for the development of scalable aligned artificial intelligence (AI) from first principle descriptions of natural intelligence. In brief, a possible path toward scalable aligned AI rests upon enabling artificial agents to learn a good model of the world that includes a good model of our preferences. For this, the main objective is creating agents that learn to represent the world and other agents' world models; a problem that falls under structure learning (a.k.a. causal representation learning). We expose the structure learning and alignment problems with this goal in mind, as well as principles to guide us forward, synthesizing various ideas across mathematics, statistics, and cognitive science. 1) We discuss the essential role of core knowledge, information geometry and model reduction in structure learning, and suggest core structural modules to learn a wide range of naturalistic worlds. 2) We outline a way toward aligned agents through structure learning and theory of mind. As an illustrative example, we mathematically sketch Asimov's Laws of Robotics, which prescribe agents to act cautiously to minimize the ill-being of other agents. We supplement this example by proposing refined approaches to alignment. These observations may guide the development of artificial intelligence in helping to scale existing -- or design new -- aligned structure learning systems.

Fine-tuning modern computer vision models requires accurately labeled data for which the ground truth may not exist, but a set of multiple labels can be obtained from labelers of variable accuracy. We tie the notion of label quality to confidence in labeler accuracy and show that, when prior estimates of labeler accuracy are available, using a simple naive-Bayes model to estimate the true labels allows us to label more data on a fixed budget without compromising label or fine-tuning quality. We present experiments on a dataset of industrial images that demonstrates that our method, called Ground Truth Extension (GTX), enables fine-tuning ML models using fewer human labels.

Diffusion models have become increasingly popular for generative modeling due to their ability to generate high-quality samples. This has unlocked exciting new possibilities for solving inverse problems, especially in image restoration and reconstruction, by treating diffusion models as unsupervised priors. This survey provides a comprehensive overview of methods that utilize pre-trained diffusion models to solve inverse problems without requiring further training. We introduce taxonomies to categorize these methods based on both the problems they address and the techniques they employ. We analyze the connections between different approaches, offering insights into their practical implementation and highlighting important considerations. We further discuss specific challenges and potential solutions associated with using latent diffusion models for inverse problems. This work aims to be a valuable resource for those interested in learning about the intersection of diffusion models and inverse problems.

For individuals and families, access to affordable credit is essential as protection against financial volatility, financing and education, pursuing business opportunities, and building equity. From the lender's perspective, there is a delicate balance between improving access to credit and higher costs due to defaults on payments. Creating responsible credit concession models requires maintaining this balance [Kozodoi et al., 2022] while ensuring fair outcomes across different groups of individuals, improving access, and helping applicants understand factors that influence rejection so that they can take action to improve their credit potential. Credit concession models are created using a variety of data, such as employment history (for example, occupation and income), demographic data (such as age, marital status, and education), and financial data (for example, checking account balance, credit card usage, and bill payment history). Given these features, models such as logistic regression, gradient boosting, and decision trees can be trained to predict whether a new customer will default on a loan over a period of time [Louzada et al., 2016].

Conflicting sensor measurements pose a huge problem for the environment representation of an autonomous robot. Therefore, in this paper, we address the self-assessment of an evidential grid map in which data from conflicting LiDAR sensor measurements are fused, followed by methods for robust motion planning under these circumstances. First, conflicting measurements aggregated in Subjective-Logic-based evidential grid maps are classified. Then, a self-assessment framework evaluates these conflicts and estimates their severity for the overall system by calculating a degradation score. This enables the detection of calibration errors and insufficient sensor setups. In contrast to other motion planning approaches, the information gained from the evidential grid maps is further used inside our proposed path-planning algorithm. Here, the impact of conflicting measurements on the current motion plan is evaluated, and a robust and curious path-planning strategy is derived to plan paths under the influence of conflicting data. This ensures that the system integrity is maintained in severely degraded environment representations which can prevent the unnecessary abortion of planning tasks.

Although language model scores are often treated as probabilities, their reliability as probability estimators has mainly been studied through calibration, overlooking other aspects. In particular, it is unclear whether language models produce the same value for different ways of assigning joint probabilities to word spans. Our work introduces a novel framework, ConTestS (Consistency Testing over Spans), involving statistical tests to assess score consistency across interchangeable completion and conditioning orders. We conduct experiments on post-release real and synthetic data to eliminate training effects. Our findings reveal that both Masked Language Models (MLMs) and autoregressive models exhibit inconsistent predictions, with autoregressive models showing larger discrepancies. Larger MLMs tend to produce more consistent predictions, while autoregressive models show the opposite trend. Moreover, for both model types, prediction entropies offer insights into the true word span likelihood and therefore can aid in selecting optimal decoding strategies. The inconsistencies revealed by our analysis, as well their connection to prediction entropies and differences between model types, can serve as useful guides for future research on addressing these limitations.

Communication between humans and robots (or virtual agents) is essential for interaction and often inspired by human communication, which uses gestures, facial expressions, gaze direction, and other explicit and implicit means. This work presents an interaction experiment where humans and virtual agents interact through explicit (gestures, manual entries using mouse and keyboard, voice, sound, and information on screen) and implicit (gaze direction, location, facial expressions, and raise of eyebrows) communication to evaluate the effect of mixed explicit-implicit communication against purely explicit communication. Results obtained using Bayesian parameter estimation show that the number of errors and task execution time did not significantly change when mixed explicit and implicit communications were used, and neither the perceived efficiency of the interaction. In contrast, acceptance, sociability, and transparency of the virtual agent increased when using mixed communication modalities (88.3%, 92%, and 92.9% of the effect size posterior distribution of each variable, respectively, were above the upper limit of the region of practical equivalence). This suggests that task-related measures, such as time, number of errors, and perceived efficiency of the interaction, have not been influenced by the communication type in our particular experiment. However, the improvement of subjective measures related to the virtual agent, such as acceptance, sociability, and transparency, suggests that humans are more receptive to mixed explicit and implicit communications.

When solving inverse problems, it is increasingly popular to use pre-trained diffusion models as plug-and-play priors. This framework can accommodate different forward models without re-training while preserving the generative capability of diffusion models. Despite their success in many imaging inverse problems, most existing methods rely on privileged information such as derivative, pseudo-inverse, or full knowledge about the forward model. This reliance poses a substantial limitation that restricts their use in a wide range of problems where such information is unavailable, such as in many scientific applications. To address this issue, we propose Ensemble Kalman Diffusion Guidance (EnKG) for diffusion models, a derivative-free approach that can solve inverse problems by only accessing forward model evaluations and a pre-trained diffusion model prior. We study the empirical effectiveness of our method across various inverse problems, including scientific settings such as inferring fluid flows and astronomical objects, which are highly non-linear inverse problems that often only permit black-box access to the forward model.

Computational models are invaluable in capturing the complexities of real-world biological processes. Yet, the selection of appropriate algorithms for inference tasks, especially when dealing with real-world observational data, remains a challenging and underexplored area. This gap has spurred the development of various parameter estimation algorithms, particularly within the realm of Simulation-Based Inference (SBI), such as neural and statistical SBI methods. Limited research exists on how to make informed choices on SBI methods when faced with real-world data, which often results in some form of model misspecification. In this paper, we provide comprehensive guidelines for deciding between SBI approaches for complex biological models. We apply the guidelines to two agent-based models that describe cellular dynamics using real-world data. Our study unveils a critical insight: while neural SBI methods demand significantly fewer simulations for inference results, they tend to yield biased estimations, a trend persistent even with robust variants of these algorithms. On the other hand, the accuracy of statistical SBI methods enhances substantially as the number of simulations increases. This finding suggests that, given a sufficient computational budget, statistical SBI can surpass neural SBI in performance. Our results not only shed light on the efficacy of different SBI methodologies in real-world scenarios but also suggest potential avenues for enhancing neural SBI approaches. This study is poised to be a useful resource for computational biologists navigating the intricate landscape of SBI in biological modeling.

Neural backdoors represent insidious cybersecurity loopholes that render learning machinery vulnerable to unauthorised manipulations, potentially enabling the weaponisation of artificial intelligence with catastrophic consequences. A backdoor attack involves the clandestine infiltration of a trigger during the learning process, metaphorically analogous to hypnopaedia, where ideas are implanted into a subject's subconscious mind under the state of hypnosis or unconsciousness. When activated by a sensory stimulus, the trigger evokes conditioned reflex that directs a machine to mount a predetermined response. In this study, we propose a cybernetic framework for constant surveillance of backdoors threats, driven by the dynamic nature of untrustworthy data sources. We develop a self-aware unlearning mechanism to autonomously detach a machine's behaviour from the backdoor trigger. Through reverse engineering and statistical inference, we detect deceptive patterns and estimate the likelihood of backdoor infection. We employ model inversion to elicit artificial mental imagery, using stochastic processes to disrupt optimisation pathways and avoid convergent but potentially flawed patterns. This is followed by hypothesis analysis, which estimates the likelihood of each potentially malicious pattern being the true trigger and infers the probability of infection. The primary objective of this study is to maintain a stable state of equilibrium between knowledge fidelity and backdoor vulnerability.