Large and diverse datasets have been the cornerstones of many impressive advancements in artificial intelligence. Intelligent creatures, however, learn by interacting with the environment, which changes the input sensory signals and the state of the environment. In this work, we aim to bring the best of both worlds and propose an algorithm that exhibits an exploratory behavior whilst it utilizes large diverse datasets. Our key idea is to leverage deep generative models that are pretrained on static datasets and introduce a dynamic model in the latent space. The transition dynamics simply mixes an action and a random sampled latent.

Over the course of an eight-minute interview, Elon Musk touted his numerous businesses and vision of a "Star Trek future" while telling the crowd that his Tesla Optimus robots had performed a dance for Donald Trump and the crown prince of Saudi Arabia, Mohammed bin Salman, to the tune of YMCA. He also announced that Starlink, his satellite internet company, had struck a deal for use in Saudi Arabia for maritime and aviation usage; looking to the near future, he expressed his desire to bring Tesla's self-driving robotaxis to the country. "We could not be more appreciative of having a lifetime partner and a friend like you, Elon, to the Kingdom," Saudi Arabia's minister of communications and IT, Abdullah Alswaha, told Musk. Although Musk has pivoted away from his role as de facto leader of the so-called "department of government efficiency" and moved out of the White House, the Saudi summit showed how he is still retaining his proximity to the US president and international influence. As Musk returns to his businesses as his primary focus, he is still primed to reap the rewards of his connections and political sway over Trump.

In reinforcement learning (RL), it is easier to solve a task if given a good representation. While deep RL should automatically acquire such good representations, prior work often finds that learning representations in an end-to-end fashion is unstable and instead equip RL algorithms with additional representation learning parts (e.g., auxiliary losses, data augmentation). How can we design RL algorithms that directly acquire good representations? In this paper, instead of adding representation learning parts to an existing RL algorithm, we show (contrastive) representation learning methods are already RL algorithms in their own right. To do this, we build upon prior work and apply contrastive representation learning to action-labeled trajectories, in such a way that the (inner product of) learned representations exactly corresponds to a goal-conditioned value function.

Despite its global prominence, and years of investment from the tech industry's loudest voices and biggest pocketbooks, AI still has a diversity problem. Filling an increasingly worrisome gap created by the tech's creators and evangelists, diversity-based organizations have been trying to tackle that issue on their own. Black Girls Code for example -- which offers tech skill building for Black girls and other historically underrecognized groups -- has been leaning more heavily into AI as part of its tech preparedness and training curriculum, including creating the brand new position of AI Expert-in-Residence to oversee a more thoughtful approach to teaching about AI. "Most AI is built in environments that prioritize profit over people, which means bias gets baked in and the same communities left out of past tech waves are now at risk of being harmed again. It's not enough to teach people to use AI, we have to teach them to be thoughtful about the tools that they use," Black Girls Code CEO Cristina Mancini tells Mashable. What values does it reflect?

Hannah, a former teacher, joins'Fox & Friends' to explain why she left the classroom, saying AI tools are making it difficult to teach. A former high school English teacher went viral this week after posting a candid video on social media announcing she was quitting the teaching profession because of how technology was "ruining" education. In her video, which reached over 1 million views on TikTok, Hannah explained how AI tools have made teaching more difficult because students rely on technology to do the work for them and are unmotivated to put in effort themselves. She said that kids do not know how to read because of read-aloud tools, and have short attention spans because of the "high stimulation" of social media. "They want to use [technology] for entertainment. They don't want to use it for education," she said in a TikTok video which reached over 1 million views.

A chatbot developed by Elon Musk's multibillion-dollar artificial intelligence startup xAI appeared to be suffering from a glitch Wednesday when it repeatedly brought up white genocide in South Africa in response to user queries about unrelated topics on X. Grok, which competes with other chatbots like OpenAI's ChatGPT, is directly integrated into the social media platform that Musk also owns. Numerous examples of the phenomenon could be found by searching the official Grok profile for posts containing the term "boer," a word used to refer to people from South Africa of "Dutch, German, or Huguenot descent." It is sometimes used by Black South Africans as a pejorative against white Afrikaners, or people associated with the apartheid regime. In response to topics ranging from streaming platform HBO Max's name change to Medicaid cuts proposed by US lawmakers, the chatbot often seemed to initially stay on topic before veering back to white genocide in South Africa, completely unprompted. When asked to confirm the salary of Toronto Blue Jays player Max Scherzer, for example, the generative artificial intelligence chatbot launched into an explanation of white genocide and a controversial South African anti-apartheid song.

We study differentially private (DP) algorithms for smooth stochastic minimax optimization, with stochastic minimization as a byproduct. The holy grail of these settings is to guarantee the optimal trade-off between the privacy and the excess population loss, using an algorithm with a linear time-complexity in the number of training samples. We provide a general framework for solving differentially private stochastic minimax optimization (DP-SMO) problems, which enables the practitioners to bring their own base optimization algorithm and use it as a black-box to obtain the near-optimal privacy-loss trade-off. Our framework is inspired from the recently proposed Phased-ERM method [22] for nonsmooth differentially private stochastic convex optimization (DP-SCO), which exploits the stability of the empirical risk minimization (ERM) for the privacy guarantee. The flexibility of our approach enables us to sidestep the requirement that the base algorithm needs to have bounded sensitivity, and allows the use of sophisticated variance-reduced accelerated methods to achieve near-linear time-complexity.

Waymo, the autonomous ride-hailing company that launched its services in Los Angeles late last year, is recalling more than 1,200 vehicles due to a software defect, the National Highway Traffic Safety Assn. said Wednesday. The recall comes after a series of minor crashes with gates, chains and other obstacles in the road that did not result in any injuries, the Mountain View, Calif.-based company said in a filing with the NHTSA. The recall applies to 1,212 driverless vehicles operating on Waymo's fifth-generation automated driving software. Waymo released a software update to resolve the issue, and that update has already been rolled out in all affected vehicles, the recall notice said. The company operates more than 1,500 vehicles across Los Angeles, San Francisco, Phoenix and Austin.

The success of deep learning comes from its ability to capture the hierarchical structure of data by learning high-level representations defined in terms of low-level ones. In this paper we explore self-supervised learning of hierarchical representations of speech by applying multiple levels of Contrastive Predictive Coding (CPC). We observe that simply stacking two CPC models does not yield significant improvements over single-level architectures. Inspired by the fact that speech is often described as a sequence of discrete units unevenly distributed in time, we propose a model in which the output of a low-level CPC module is non-uniformly downsampled to directly minimize the loss of a high-level CPC module. The latter is designed to also enforce a prior of separability and discreteness in its representations by enforcing dissimilarity of successive high-level representations through focused negative sampling, and by quantization of the prediction targets.

In academic recruitment settings, including faculty hiring and PhD admissions, committees aim to maximize the overall quality of recruited candidates, but there is uncertainty about whether a candidate would accept an offer if given one. Previous work has considered algorithms that make offers sequentially and are subject to a hard budget constraint. We argue that these modeling choices may be inconsistent with the practice of academic recruitment. Instead, we restrict ourselves to a single batch of offers, and we treat the target number of positions as a soft constraint, so we risk overshooting or undershooting the target. Specifically, our objective is to select a subset of candidates that maximizes the overall expected value associated with candidates who accept, minus an expected penalty for deviating from the target.