Before deploying outputs from foundation models in high-stakes tasks, it is imperative to ensure that they align with human values.For instance, in radiology report generation, reports generated by a vision-language model must align with human evaluations before their use in medical decision-making. This paper presents Conformal Alignment, a general framework for identifying units whose outputs meet a user-specified alignment criterion. It is guaranteed that on average, a prescribed fraction of selected units indeed meet the alignment criterion, regardless of the foundation model or the data distribution. Given any pre-trained model and new units with model-generated outputs, Conformal Alignment leverages a set of reference data with ground-truth alignment status to train an alignment predictor. It then selects new units whose predicted alignment scores surpass a data-dependent threshold, certifying their corresponding outputs as trustworthy. Through applications to question answering and radiology report generation, we demonstrate that our method is able to accurately identify units with trustworthy outputs via lightweight training over a moderate amount of reference data. En route, we investigate the informativeness of various features in alignment prediction and combine them with standard models to construct the alignment predictor.

Before deploying outputs from foundation models in high-stakes tasks, it is imperative to ensure that they align with human values.For instance, in radiology report generation, reports generated by a vision-language model must align with human evaluations before their use in medical decision-making. This paper presents Conformal Alignment, a general framework for identifying units whose outputs meet a user-specified alignment criterion. It is guaranteed that on average, a prescribed fraction of selected units indeed meet the alignment criterion, regardless of the foundation model or the data distribution. Given any pre-trained model and new units with model-generated outputs, Conformal Alignment leverages a set of reference data with ground-truth alignment status to train an alignment predictor. It then selects new units whose predicted alignment scores surpass a data-dependent threshold, certifying their corresponding outputs as trustworthy. Through applications to question answering and radiology report generation, we demonstrate that our method is able to accurately identify units with trustworthy outputs via lightweight training over a moderate amount of reference data.

Argentina's security forces have announced plans to use artificial intelligence to "predict future crimes" in a move experts have warned could threaten citizens' rights. The country's far-right president Javier Milei this week created the Artificial Intelligence Applied to Security Unit, which the legislation says will use "machine-learning algorithms to analyse historical crime data to predict future crimes". It is also expected to deploy facial recognition software to identify "wanted persons", patrol social media, and analyse real-time security camera footage to detect suspicious activities. While the ministry of security has said the new unit will help to "detect potential threats, identify movements of criminal groups or anticipate disturbances", the Minority Report-esque resolution has sent alarm bells ringing among human rights organisations. Experts fear that certain groups of society could be overly scrutinised by the technology, and have also raised concerns over who – and how many security forces – will be able to access the information.

This paper explores a federated learning approach that automatically selects the number of latent processes in multi-output Gaussian processes (MGPs). The MGP has seen great success as a transfer learning tool when data is generated from multiple sources/units/entities. A common approach in MGPs to transfer knowledge across units involves gathering all data from each unit to a central server and extracting common independent latent processes to express each unit as a linear combination of the shared latent patterns. However, this approach poses key challenges in (i) determining the adequate number of latent processes and (ii) relying on centralized learning which leads to potential privacy risks and significant computational burdens on the central server. To address these issues, we propose a hierarchical model that places spike-and-slab priors on the coefficients of each latent process. These priors help automatically select only needed latent processes by shrinking the coefficients of unnecessary ones to zero. To estimate the model while avoiding the drawbacks of centralized learning, we propose a variational inference-based approach, that formulates model inference as an optimization problem compatible with federated settings. We then design a federated learning algorithm that allows units to jointly select and infer the common latent processes without sharing their data. We also discuss an efficient learning approach for a new unit within our proposed federated framework. Simulation and case studies on Li-ion battery degradation and air temperature data demonstrate the advantageous features of our proposed approach.

Recent literature has explored various ways to improve soft sensors using learning algorithms with transferability. Broadly put, the performance of a soft sensor may be strengthened when it is learned by solving multiple tasks. The usefulness of transferability depends on how strongly related the devised learning tasks are. A particularly relevant case for transferability, is when a soft sensor is to be developed for a process of which there are many realizations, e.g. system or device with many implementations from which data is available. Then, each realization presents a soft sensor learning task, and it is reasonable to expect that the different tasks are strongly related. Applying transferability in this setting leads to what we call multi-unit soft sensing, where a soft sensor models a process by learning from data from all of its realizations. This paper explores the learning abilities of a multi-unit soft sensor, which is formulated as a hierarchical model and implemented using a deep neural network. In particular, we investigate how well the soft sensor generalizes as the number of units increase. Using a large industrial dataset, we demonstrate that, when the soft sensor is learned from a sufficient number of tasks, it permits few-shot learning on data from new units. Surprisingly, regarding the difficulty of the task, few-shot learning on 1-3 data points often leads to a high performance on new units.

In fields such as medicine and drug discovery, the ultimate goal of a classification is not to guess a class, but to choose the optimal course of action among a set of possible ones, usually not in one-one correspondence with the set of classes. This decision-theoretic problem requires sensible probabilities for the classes. Probabilities conditional on the features are computationally almost impossible to find in many important cases. The main idea of the present work is to calculate probabilities conditional not on the features, but on the trained classifier's output. This calculation is cheap, needs to be made only once, and provides an output-to-probability "transducer" that can be applied to all future outputs of the classifier. In conjunction with problem-dependent utilities, the probabilities of the transducer allow us to find the optimal choice among the classes or among a set of more general decisions, by means of expected-utility maximization. This idea is demonstrated in a simplified drug-discovery problem with a highly imbalanced dataset. The transducer and utility maximization together always lead to improved results, sometimes close to theoretical maximum, for all sets of problem-dependent utilities. The one-time-only calculation of the transducer also provides, automatically: (i) a quantification of the uncertainty about the transducer itself; (ii) the expected utility of the augmented algorithm (including its uncertainty), which can be used for algorithm selection; (iii) the possibility of using the algorithm in a "generative mode", useful if the training dataset is biased.

Night Market, a unit launched by Horizon Media three years ago to specialize in the burgeoning direct-to-consumer (D2C) retail marketplace, this morning announced its own new unit -- Neon -- which utilizes proprietary artificial intelligence that buys media smarter than people can. "[Neon is] designed to increase advertisers' revenue by 20 % when planning and buying retail media to achieve their maximum revenue outcomes," the agency boasts in a statement unveiling the new unit, adding: "Neon starts with the sale in mind using automated predictive analytics to make investment decisions across retailers." Specifically, Horizon said the platform's "computing power forecasts to the tactical level and optimizes in-market spend in real-time at the SKU level, citing these explicit capabilities:

Real-Time Strategy (RTS) game unit generation is an unexplored area of Procedural Content Generation (PCG) research, which leaves the question of how to automatically generate interesting and balanced units unanswered. Creating unique and balanced units can be a difficult task when designing an RTS game, even for humans. Having an automated method of designing units could help developers speed up the creation process as well as find new ideas. In this work we propose a method of generating balanced and useful RTS units. We draw on Search-Based PCG and a fitness function based on Monte Carlo Tree Search (MCTS). We present ten units generated by our system designed to be used in the game microRTS, as well as results demonstrating that these units are unique, useful, and balanced.

Ford CEO Jim Farley has made quite a few changes to the way the automaker does things since his arrival. Now, Ford is doing something similar with its autonomous technology divisions. Autonomous Tech Gets A New Home According to a report by Bloomberg, Ford is looking to accelerate the development of its autonomous technology by forming a new division called Ford Next. Apparently, Farley formed Ford Next last year and put Franck Louis-Victor, a specialist in new businesses from Renault SA, at the head. The division includes the automaker's stake in Argo Ai, an autonomous startup, as well as Ford's self-driving unit that's called Ford Autonomous Vehicles LLC.

LAS VEGAS – Sony Group Corp. will set up a new unit in the spring for electric vehicles as it explores the possibility of launching the vehicles commercially, its CEO said Tuesday. Speaking at a media preview ahead of the annual Consumer Electronics Show in Las Vegas, Sony Group CEO Kenichiro Yoshida said the new unit, Sony Mobility Inc., will aim to make the best use of artificial intelligence and robotics technology for EV development. At the same event in 2020, Sony unveiled a prototype EV, the Vision-S, which is equipped with technology for autonomous driving and aimed at enhancing the safety and comfort of mobility. The announcement came at a time when global competition has been intensifying over the development of EVs, with many automakers shifting to such vehicles for carbon emission reduction. At this year's preview, Yoshida also unveiled an SUV prototype of its Vision-S EV, which features seat speakers that create a three-dimensional sound field.