|
Amazon announced that it has created its own quantum computing chip, joining Microsoft and Google in a push to take this potentially transformative technology from the theoretical to the practical. Ocelot is a prototype that's intended to test the effectiveness of Amazon Web Services' quantum error correction architecture. Compared with other chip methods, the company claims Ocelot can reduce the cost of implementing quantum error correction by up to 90 percent. Quantum computing could solve complicated problems exponentially faster than standard computers by using quantum bits, or qubits, rather than traditional bits that store a computer's information as 1s and 0s. Rather than representing only a 1 or a 0, qubits can represent a proportion of both 1 and 0 at the same time. Ocelot takes this a step farther with its use of "cat qubits," named for the famous Schrödinger's cat thought experiment, which can "intrinsically suppress certain forms of errors," according to Amazon. Error rate is one of the key limitations of current quantum computing, because the qubits are so sensitive to minute changes in their environments. Electromagnetic interference from a WiFi network can be enough to disturb a qubit and cause it to make mistakes. Adding more qubits to a chip means faster calculations, but also more mistakes. Ocelot consists of five data qubits (the cat qubits), five "buffer circuits" to stabilize them and four qubits for detecting errors on the cat qubits. "We selected our qubit and architecture with quantum error correction as the top requirement," said Oskar Painter, director of quantum hardware at AWS. "We believe that if we're going to make practical quantum computers, quantum error correction needs to come first." Google claimed that its Willow chip, announced in December, was able to reduce errors as more qubits were added. Ocelot is another step toward reducing errors and making useful quantum computers a reality.This article originally appeared on Engadget at https://www.engadget.com/computing/amazon-joins-the-quantum-computing-race-with-a-chip-designed-for-error-correction-201501075.html?src=rss
Category:
Marketing and Advertising
In what has already been a busy past few days for new model releases, OpenAI is capping off the week with a research preview of GPT-4.5. The company is touting the new system as its largest and best model for chat yet. In early testing, OpenAI says people found GPT-4.5 to be a more natural conversationalist, with the ability to convey warmth and display a kind of emotional intelligence. In one example shared by OpenAI, a person tells ChatGPT they're going through a hard time after failing a test. Where the company's previous models, including GPT-4o and o3-mini, might commiserate with the individual before offering a long list of unsolicited advice, GPT-4.5 takes a different tact. "Want to talk about what happened, or do you just need a distraction? I'm here either way," the chatbot says when powered by GPT-4.5. The gains shown by GPT-4.5 are the result of advancements OpenAI made in unsupervised learning. With unsupervised learning, a machine learning algorithm is given an unlabeled data set and left to its own devices to find patterns and insights. GPT-4.5 doesn't "think" like the company's state-of-the-art reasoning models, but in training the new model OpenAI made architectural enhancements and gave it access to more data and compute power. "The result is a model that has broader knowledge and a deeper understanding of the world, leading to reduced hallucinations," the company says. Speaking of reduced hallucinations, OpenAI measured how much better GPT-4.5 in that regard. When put through SimpleQA, an OpenAI-designed benchmark that tests large language models on their ability to answer "straightforward but challenging knowledge questions," GPT-4.5 beat out o3-mini, GPT-4o and even o1 with a hallucination rate of 37.1 percent. Obviously, the new model doesn't solve the problem of AI hallucinations altogether, but it is a step in the right direction. Despite its relative strengths over GPT-4o and o3-mini, GPT-4.5 isn't a direct replacement for those models. Compared to OpenAI's reasoning systems, GPT-4.5 is "a more general-purpose, innately smarter model." Additionally, it's not natively multimodal like GPT-4o, meaning it doesn't work with features like Voice Mode, video or screensharing. Its also "a very large and compute-intensive model." It's best to think of GPT-4.5 as a stepping stone to systems OpenAI plans to offer in the future. In fact, Sam Altman said as much earlier this month when he shared the company's roadmap, noting GPT-4.5 would be "our last non-chain-of-thought model" referring to the fact that the new system doesn't solve problems by tackling them step by step like OpenAI's reasoning models do. Its successor, GPT-5, will likely integrate many of OpenAI's latest technologies, including its frontier o3 model. OpenAI reiterated that today, saying it plans to bring GPT-4.5's "unique strengths, including broader knowledge, stronger intuition, and greater 'EQ,' to all users in future models." In the meantime, ChatGPT Pro subscribers can begin using GPT-4.5 starting today, with Pro and Team users slated to gain access starting next week.This article originally appeared on Engadget at https://www.engadget.com/ai/openais-new-gpt-45-model-is-a-better-more-natural-conversationalist-200035185.html?src=rss
Category:
Marketing and Advertising
Game Freak dropped a surprise trailer at todays Pokémon Presents livestream for something called Pokémon Champions. Its a dedicated battle sim, recalling the glory days of Pokémon Stadium for the N64. This means that Champions distills the franchise's formula down to just battling, with a heavy emphasis on multiplayer. To that end, the game is cross-platform, with the developer saying that players on Nintendo Switch and mobile will be able to duke it out. We dont know what kind of roster the game itself will provide, but it integrates with the cloud-storage service Pokémon Home. People should be able to pull most of their favorite Pokémon from titles like Pokémon Go and recent mainline entries like Pokémon Sword and Shield. It also works with the spin-off Pokémon Legends: Arceus. The official press release, however, does note that there will be some restrictions when it comes to availability. The gameplay should feel instantly familiar to anyone who has ever dabbled in the franchise, as the trailer description touts tried-and-true mechanics such as Pokémon types, abilities and moves. The footage even shows more niche mechanics like Mega Evolution and Terastallization, indicating that players may not be tied to the base-level Pokémon rule set. Theres no release window yet. All we know is that its currently in development. Speaking of release dates, todays livestream did reveal that Pokémon Legends: Z-A will be coming out by the end of the year for the Switch.This article originally appeared on Engadget at https://www.engadget.com/gaming/nintendo/pokemon-champions-is-all-about-the-battles-194527252.html?src=rss
Category:
Marketing and Advertising
All news |
||||||||||||||||||
|