This paper investigates how interacting agents arrive to a consensus or a polarized state. We study the opinion formation process under the effect of a global steering mechanism (GSM), which aggregates the opinion-driven stochastic agent states at the network level and feeds back to them a form of global information. We also propose a new two-layer agent-based opinion formation model, called GSM-DeGroot, that captures the coupled dynamics between agent-to-agent local interactions and the GSM's steering effect. This way, agents are subject to the effects of a DeGroot-like local opinion propagation, as well as to a wide variety of possible aggregated information that can affect their opinions, such as trending news feeds, press coverage, polls, elections, etc. Contrary to the standard DeGroot model, our model allows polarization to emerge by letting agents react to the global information in a stubborn differential way. Moreover, the introduced stochastic agent states produce event stream dynamics that can fit to real event data. We explore numerically the model dynamics to find regimes of qualitatively different behavior. We also challenge our model by fitting it to the dynamics of real topics that attracted the public attention and were recorded on Twitter. Our experiments show that the proposed model holds explanatory power, as it evidently captures real opinion formation dynamics via a relatively small set of interpretable parameters.

To improve the predictability of complex computational models in the experimentally-unknown domains, we propose a Bayesian statistical machine learning framework utilizing the Dirichlet distribution that combines results of several imperfect models. This framework can be viewed as an extension of Bayesian stacking. To illustrate the method, we study the ability of Bayesian model averaging and mixing techniques to mine nuclear masses. We show that the global and local mixtures of models reach excellent performance on both prediction accuracy and uncertainty quantification and are preferable to classical Bayesian model averaging. Additionally, our statistical analysis indicates that improving model predictions through mixing rather than mixing of corrected models leads to more robust extrapolations.

Consider learning a generative model for time-series data. The sequential setting poses a unique challenge: Not only should the generator capture the conditional dynamics of (stepwise) transitions, but its open-loop rollouts should also preserve the joint distribution of (multi-step) trajectories. On one hand, autoregressive models trained by MLE allow learning and computing explicit transition distributions, but suffer from compounding error during rollouts. On the other hand, adversarial models based on GAN training alleviate such exposure bias, but transitions are implicit and hard to assess. In this work, we study a generative framework that seeks to combine the strengths of both: Motivated by a moment-matching objective to mitigate compounding error, we optimize a local (but forward-looking) transition policy, where the reinforcement signal is provided by a global (but stepwise-decomposable) energy model trained by contrastive estimation. At training, the two components are learned cooperatively, avoiding the instabilities typical of adversarial objectives. At inference, the learned policy serves as the generator for iterative sampling, and the learned energy serves as a trajectory-level measure for evaluating sample quality. By expressly training a policy to imitate sequential behavior of time-series features in a dataset, this approach embodies "generation by imitation". Theoretically, we illustrate the correctness of this formulation and the consistency of the algorithm. Empirically, we evaluate its ability to generate predictively useful samples from real-world datasets, verifying that it performs at the standard of existing benchmarks.

Simulation-based inference (SBI) enables amortized Bayesian inference for simulators with implicit likelihoods. But when we are primarily interested in the quality of predictive simulations, or when the model cannot exactly reproduce the observed data (i.e., is misspecified), targeting the Bayesian posterior may be overly restrictive. Generalized Bayesian Inference (GBI) aims to robustify inference for (misspecified) simulator models, replacing the likelihood-function with a cost function that evaluates the goodness of parameters relative to data. However, GBI methods generally require running multiple simulations to estimate the cost function at each parameter value during inference, making the approach computationally infeasible for even moderately complex simulators. Here, we propose amortized cost estimation (ACE) for GBI to address this challenge: We train a neural network to approximate the cost function, which we define as the expected distance between simulations produced by a parameter and observed data. The trained network can then be used with MCMC to infer GBI posteriors for any observation without running additional simulations. We show that, on several benchmark tasks, ACE accurately predicts cost and provides predictive simulations that are closer to synthetic observations than other SBI methods, especially for misspecified simulators. Finally, we apply ACE to infer parameters of the Hodgkin-Huxley model given real intracellular recordings from the Allen Cell Types Database. ACE identifies better data-matching parameters while being an order of magnitude more simulation-efficient than a standard SBI method. In summary, ACE combines the strengths of SBI methods and GBI to perform robust and simulation-amortized inference for scientific simulators.

The White House issued draft rules today that would require federal agencies to evaluate and constantly monitor algorithms used in health care, law enforcement, and housing for potential discrimination or other harmful effects on human rights. Once in effect, the rules could force changes in US government activity dependent on AI, such as the FBI's use of face recognition technology, which has been criticized for not taking steps called for by Congress to protect civil liberties. The new rules would require government agencies to assess existing algorithms by August 2024 and stop using any that don't comply. "If the benefits do not meaningfully outweigh the risks, agencies should not use the AI," the memo says. But the draft memo carves out an exemption for models that deal with national security and allows agencies to effectively issue themselves waivers if ending use of an AI model "would create an unacceptable impediment to critical agency operations."

President Biden this week signed a sweeping Executive Order on artificial intelligence that seeks to tackle threats posed by the technology, but some experts say the regulation has left questions unanswered about how it could work in practice. The order tasks agencies with rethinking their approach to AI and aims to address threats relating to national security, competition and consumer privacy, while promoting innovation, competition, and the use of AI for public services. One of the most significant elements of the order is the requirement for companies developing the most powerful AI models to disclose the results of safety tests. On Tuesday, Secretary of Commerce Gina Raimondo told CNBC that under the Executive Order "the President directs the Commerce Department to require companies to tell us: what are the safety precautions they're putting in place and to allow us to judge whether that's enough. And we plan to hold these companies accountable."

The global AI safety summit opened at Bletchley Park on Wednesday with a landmark declaration from countries including the UK, US, EU and China that the technology poses a potentially catastrophic risk to humanity. The so-called Bletchley declaration said: "There is potential for serious, even catastrophic, harm, either deliberate or unintentional, stemming from the most significant capabilities of these AI models." Here are some of the interventions from political and tech industry figures – as well as King Charles – on the day. The world's richest man and Tesla chief executive described AI as a threat to humanity. Musk, who co-founded the ChatGPT developer OpenAI, has launched a new venture called xAI and is attending both days of the summit, which is being held about 50 miles from London at the site which played host to top-secret codebreakers during the second world war.

Yemen's Houthi militia claimed an attempted attack on southern Israel on Tuesday, saying it had launched a "large batch" of ballistic and cruise missiles as well as drones toward Israeli targets. The Iran-backed militia carried out the attempted assault in response to what it called "brutal Israeli-American aggression" in Gaza, the Houthi military spokesman, Yahya Sarea, said on the social media platform X. Mr. Sarea said the attack was the third operation conducted by the Houthis "in support of our persecuted brothers in Palestine," and threatened further missile and drone assaults. The Times could not independently verify the Houthi claims. On Tuesday, the Israeli military said its aerial defense system had intercepted a surface-to-surface missile fired toward Israel "from the area of the Red Sea."

The challenge of regulating artificial intelligence is sometimes compared to the management of nuclear energy: there are valuable civil applications alongside terrifying military ones, and a credible risk of existential calamity if it all goes wrong. But nuclear weapons are expensive and hard to acquire. By contrast, AI can distribute awesome power at relatively low cost. This adds unprecedented complexity to the task facing attenders at an AI safety summit that Rishi Sunak is hosting this week at Bletchley Park. The prime minister wants to position the UK as a global leader in the field.

Speaking ahead of the event in London, US Vice President Kamala Harris said that world leaders "must address the full spectrum of AI risks to humanity" and listed examples of faulty algorithms in healthcare, the use of AI in making "deepfakes", misinformation and biased facial recognition.