Is Silicon Valley Still the Tech Capital? On this special episode of recorded in front of a live audience in San Francisco, our hosts discuss Silicon Valley's history and future. All products featured on WIRED are independently selected by our editors. However, we may receive compensation from retailers and/or from purchases of products through these links. Earlier this month, we took the show to San Francisco for a live recording in front of a great audience at KQED's The Commons. WIRED's Lauren Goode, Katie Drummond, and Jason Kehe asked themselves and answered a perennial question: Is Silicon Valley still the tech capital of the world? Plus, they put themselves to the test with a new game and some questions from the audience. Write to us at uncannyvalley@wired.com . You can always listen to this week's podcast through the audio player on this page, but if you want to subscribe for free to get every episode, here's how: If you're on an iPhone or iPad, open the app called Podcasts, or just tap this link . This week, we wanted to share with you the live show that we hosted in partnership with KQED earlier this month. Our global editorial director, Katie Drummond, had a really fun and sharp conversation with Jack Conte, the CEO of Patreon, about what it takes to make authentic work in the era of AI and influencers. Then my cohosts, WIRED's Lauren Goode and Jason Kehe, joined the stage for a special roundtable discussion. I sadly couldn't make it, but I feel very thankful for everyone who came through that night. I hope you enjoy it. I just want to say before we officially get started, Michael Colore, who a lot of you know and love, our beloved "Snackfight," could not be here tonight because he is weathering Covid. He said we can share that, but we are so glad to have Jason here in his stead. Actually, how do we say that? I suppose you could say that in Mike's stead-- I forced Jason to come do it tonight. Bully was the word I was using. It's going to be great. We're so thrilled, and he's going to have plenty of spicy takes. So without any further ado, welcome again everyone to our first live edition of the Roundtable here in San Francisco with our partners, KQED.

Oregon authorities recaptured Ty Anthony Sage after he was accidentally freed on bail despite a judge's order denying bail in the 2021 killing of Lowgunn Ivey.

Can Ukraine restore its pre-war borders? Is Russia testing NATO with aerial incursions in Europe? EU to develop'drone wall' amid regional airspace violations European Union defence ministers have agreed to develop a "drone wall" along their borders with Russia and Ukraine to detect, track and intercept violations of their airspace, according to the bloc's defence chief. Friday's announcement comes after rogue drones entered Polish airspace on September 10, rattling eastern EU members. Although Danish authorities have not concluded their investigations, Frederiksen stressed that Russia was currently the primary threat to European security. The Kremlin has denied any involvement in the drone incidents in Poland and Denmark.

From Montenegro to northern California, the tech elite dream of building cities where they make the rules. Is this, finally, their moment? The shirtless man in the golden mask and cape has plans to lead his own country one day. There is no location yet, but it will be a crypto-and AI-powered paradise of medical experimentation, filled with people who want to "make death optional," he says. For now, though, he's leading a sparsely attended rave on the second floor of a San Francisco office building. A DJ is spinning at one end of an open room. A handful of people sway and jump on the space cleared out as a dance floor. At a nearby table, coffee is available with many alternative milks.

The Take Ahmed al-Sharaa's high-stakes bid to remake Syria For the first time in nearly 60 years, a Syrian leader speaks at the United Nations. It was a symbolic moment for a nation long-isolated from the international stage. President Ahmed al-Sharaa says he can rebuild Syria through private investment and a deal with Israel. But how much can he concede in the name of progress, without losing Syrian public support? What is the future of US dissent in the post-Charlie Kirk era? The White House Peace Vigil is dismantled - why? How is China using AI in the classroom?

Military veteran Travis Decker's remains were found in Washington wilderness, ending a three-month manhunt after he murdered his daughters Paityn, Evelyn and Olivia.

Pre-trained transformer models have achieved remarkable success in natural language processing (NLP) and have recently become competitive alternatives to Convolution Neural Networks (CNN) and Recurrent Neural Networks (RNN) in vision and speech tasks, respectively. Due to excellent computational efficiency and scalability, transformer models can be trained on exceedingly large amounts of data; however, model sizes can grow tremendously. As high performance, large-scale, and pre-trained transformer models become available for users to download and fine-tune for customized downstream tasks, the deployment of these models becomes challenging due to the vast amount of operations and large memory footprint. To address this challenge, we introduce methods to deeply compress pre-trained transformer models across three major application domains: NLP, speech, and vision. Specifically, we quantize transformer backbones down to 4-bit and further achieve 50% fine-grained structural sparsity on pre-trained BERT, Wav2vec2.0

We prove the Fast Equilibrium Conjecture proposed by Li et al., (2020), i.e., stochastic gradient descent (SGD) on a scale-invariant loss (e.g., using networks with various normalization schemes) with learning rate \eta and weight decay factor \lambda mixes in function space in \mathcal{\tilde{O}}(\frac{1}{\lambda\eta}) steps, under two standard assumptions: (1) the noise covariance matrix is non-degenerate and (2) the minimizers of the loss form a connected, compact and analytic manifold. The analysis uses the framework of Li et al., (2021) and shows that for every T 0, the iterates of SGD with learning rate \eta and weight decay factor \lambda on the scale-invariant loss converge in distribution in \Theta\left(\eta {-1}\lambda {-1}(T \ln(\lambda/\eta))\right) iterations as \eta\lambda\to 0 while satisfying \eta \le O(\lambda)\le O(1) . Moreover, the evolution of the limiting distribution can be described by a stochastic differential equation that mixes to the same equilibrium distribution for every initialization around the manifold of minimizers as T\to\infty .

The factorization of state-action value functions for Multi-Agent Reinforcement Learning (MARL) is important. Existing studies are limited by their representation capability, sample efficiency, and approximation error. To address these challenges, we propose, ResQ, a MARL value function factorization method, which can find the optimal joint policy for any state-action value function through residual functions. ResQ masks some state-action value pairs from a joint state-action value function, which is transformed as the sum of a main function and a residual function. ResQ can be used with mean-value and stochastic-value RL.

Softmax policy gradient is a popular algorithm for policy optimization in single-agent reinforcement learning, particularly since projection is not needed for each gradient update. However, in multi-agent systems, the lack of central coordination introduces significant additional difficulties in the convergence analysis. Even for a stochastic game with identical interest, there can be multiple Nash Equilibria (NEs), which disables proof techniques that rely on the existence of a unique global optimum. Moreover, the softmax parameterization introduces non-NE policies with zero gradient, making it difficult for gradient-based algorithms in seeking NEs. In this paper, we study the finite time convergence of decentralized softmax gradient play in a special form of game, Markov Potential Games (MPGs), which includes the identical interest game as a special case. We investigate both gradient play and natural gradient play, with and without \log -barrier regularization.