Integrating large language models (LLMs) into autonomous driving has attracted significant attention with the hope of improving generalization and explainability. However, existing methods often focus on either driving or vision-language understanding but achieving both high driving performance and extensive language understanding remains challenging. In addition, the dominant approach to tackle vision-language understanding is using visual question answering. However, for autonomous driving, this is only useful if it is aligned with the action space. Otherwise, the model's answers could be inconsistent with its behavior. Therefore, we propose a model that can handle three different tasks: (1) closed-loop driving, (2) vision-language understanding, and (3) language-action alignment. Our model SimLingo is based on a vision language model (VLM) and works using only camera, excluding expensive sensors like LiDAR. SimLingo obtains state-of-the-art performance on the widely used CARLA simulator on the Bench2Drive benchmark and is the winning entry at the CARLA challenge 2024. Additionally, we achieve strong results in a wide variety of language-related tasks while maintaining high driving performance.

During waking and sleep, the brain and mind undergo a tightly linked and precisely specified set of changes in state. At the level of neurons, this process has been modeled by variations of Volterra-Lotka equations for cyclic fluctuations of brainstem cell populations. However, neural network models based upon rapidly developing knowledge ofthe specific population connectivities and their differential responses to drugs have not yet been developed. Furthermore, only the most preliminary attempts have been made to model across states. Some of our own attempts to link rapid eye movement (REM) sleep neurophysiology and dream cognition using neural network approaches are summarized in this paper.

Participatory Artificial Intelligence (PAI) has recently gained interest by researchers as means to inform the design of technology through collective's lived experience. PAI has a greater promise than that of providing useful input to developers, it can contribute to the process of democratizing the design of technology, setting the focus on what should be designed. However, in the process of PAI there existing institutional power dynamics that hinder the realization of expansive dreams and aspirations of the relevant stakeholders. In this work we propose co-design principals for AI that address institutional power dynamics focusing on Participatory AI with youth.

Imagine a collection of books--maybe millions or even billions of them--haphazardly tossed by publishers into a heaping pile in a field. Every day the pile grows exponentially. Those books are brimming with knowledge and answers. But how would a seeker find them? Lacking organization, the books are useless. This is the raw internet in all its unfiltered glory.

It's a bit of a bummer that dreams are as fascinating as they are hard and expensive to study. Famed psychologists like Sigmund Freud and Carl Jung may have made big names for themselves mining the meaning and significance of our dreams, but even today, with powerful brain-monitoring technology, it's tough to get a handle on what, exactly, is going on. Researchers, if they wait to wake up their subjects from sleep in the morning, have to contend with "rapid forgetting." A better method is to wake people up while they're dreaming, but this requires running a sleep lab, which doesn't offer that much of an advantage. The dreamers are groggy and still forgetful.

Without a doubt, the biggest questions about dreaming are all variants on this question: Why do we dream? We began studying dreaming in the early 1990s and, between the two of us, have published over 200 scientific papers on sleep and dreams. Pulling together a variety of compelling neuroscientific ideas and state-of-the-art findings in the fields of sleep and dream research, we propose a new and innovative model of why we dream. We call this model NEXTUP. It proposes that our dreams allow us to explore the brain's neural network connections in order to understand possibilities.

The crystal ball that can predict when, where, and how a customer will spend their money has a new name and it goes by Programmai. This predictive marketing software uses customer data to accurately predict customer spending for both the short and long term. Programmai can even predict the future lifetime value of a customer based on the first interaction. Applying machine learning, marketers can use predictive values from Programmani to develop marketing programs and hone acquisition costs, no longer taking stabs in the dark but instead basing these decisions on justifiable data and forecasts. London TechWatch caught up with CEO Dean Murr to find out how his previous experience at Asos led him to create Programmai, the company's experience raising during the pandemic, and recent funding round.

In the present paper, we propose a decoder-free extension of Dreamer, a leading model-based reinforcement learning (MBRL) method from pixels. Dreamer is a sample- and cost-efficient solution to robot learning, as it is used to train latent state-space models based on a variational autoencoder and to conduct policy optimization by latent trajectory imagination. However, this autoencoding based approach often causes object vanishing, in which the autoencoder fails to perceives key objects for solving control tasks, and thus significantly limiting Dreamer's potential. This work aims to relieve this Dreamer's bottleneck and enhance its performance by means of removing the decoder. For this purpose, we firstly derive a likelihood-free and InfoMax objective of contrastive learning from the evidence lower bound of Dreamer. Secondly, we incorporate two components, (i) independent linear dynamics and (ii) the random crop data augmentation, to the learning scheme so as to improve the training performance. In comparison to Dreamer and other recent model-free reinforcement learning methods, our newly devised Dreamer with InfoMax and without generative decoder (Dreaming) achieves the best scores on 5 difficult simulated robotics tasks, in which Dreamer suffers from object vanishing.

Learning to control robots directly based on images is a primary challenge in robotics. However, many existing reinforcement learning approaches require iteratively obtaining millions of samples to learn a policy which can take significant time. In this paper, we focus on the problem of learning real-world robot action policies solely based on a few random off-policy samples. We learn a realistic dreaming model that can emulate samples equivalent to a sequence of images from the actual environment, and make the agent learn action policies by interacting with the dreaming model rather than the real world. We experimentally confirm that our dreaming model can learn realistic policies that transfer to the real-world.

Michael Cook, a 30-year-old senior research fellow at the University of Falmouth, has built an AI capable of imagining new video games from scratch. Cook calls the machine Angelina, a recursive acronym that stands for "A Novel…