Ubisoft appoints new chief people officerSbastien Froidefond will "lead Ubisoft's global people strategy, driving excellence in human resources across the company" News by Vikki Blake Contributor Published on Nov. 13, 2024 Ubisoft has appointed a new chief people officer: Sbastien Froidefond.The megacorp says Froidefond will assume the role a week today, November 20, and "lead Ubisoft's global people strategy, driving excellence in human resources across the company." Froidefond will report to EVP of communications, Ccile Russeil.Froidefond's expertise "spans HR strategy, performance management, culture building and employee engagement." He joins Ubisoft following a position as human resources Americans lead at Lactalis Americas.Previously, he has worked at senior HR positions across a number of sectors, including beauty, pharma, and life sciences.Froidefond replaces Anika Grant, who left the publisher back in November 2023.Last month, over 700 Ubisoft France staff walked out on a three-day strike in a dispute over home working and pay. Ubisoft's France-based workers were called to participate in a three-day strike over a dispute around the company's return to office policy. French game workers union STJV (Syndicat des Travailleurs et Travailleuses du Jeu Vido) said the diktat to be in the office at least three days per week going forward was "the straw that broke the camel's back."

Casual studio Funstop Games secures $5 million in Series A fundingFunding will be used to "ramp up" development efforts and expand portfolioImage credit: Funstop Games News by Vikki Blake Contributor Published on Nov. 13, 2024 Casual gaming studio Funstop Games has secured $5 million in its Series A funding.As reported by The Economic Times, the funding round - led by 360 One Asset - will be used to "ramp up" Funstop's development efforts, expand its portfolio, and build its ad-tech systems.With 360 One Asset coming onboard, we aim to be truly global with rapid expansion across the genres and doubling down on the India growth story," said co-founder Prashanth Vijay."The 360 One Asset team would bring in a [sic] depth understanding about the gaming ecosystem, a playbook on global expansion, and most importantly, the know-how on scaling up rapidly, making them ideal partners for us at this stage.India's gaming industry is on the cusp of global recognition, and Funstop Games is here to drive that momentum. This investment gives us the resources to scale our operations, expand our portfolio, and bring uniquely Indian storytelling to a global audience," added additional co-founder Tamesh Sharma.

Justin Carter, Contributing EditorNovember 13, 20242 Min ReadImage via Microsoft.At a GlanceSpencer believes Xbox still needs to enter the mobile and Asian markets, and is determined to make both happen.Microsoft has made plenty of acquisitions in recent years, and its games division head Phil Spencer believes there are still more to come.Speaking with Bloomberg, Spencer said the company is still looking at third-party studios for potential purchase. No one specific developer was cited, but he did say large-scale purchases like Activision Blizzard and Bethesda were on pause, at least for now. By his own admission, there was "a lot of time" spent acquiring the Call of Duty publisher, which became complicated by local and international regulators initially stepping in to block it.The acquisition went through last year, and resulted in Call of Duty: Black Ops 6 becoming a big success as a day one Xbox Game Pass title. However, Xbox has also seen its share of controversial layoffs and closures across its subsidiaries, with one of the most recent being 650 roles eliminated this past September.While Spencer stated no acquisitions were "imminent," he expressed interest in both non-US and mobile studios. The former interests him because of "geographic diversity:" previously, Xbox attempted to get Chinese hits like Genshin Impact on the platformwhich it will, next weekand he told Bloomberg there's a "real opportunity to partner with creative teams in China for global [reach]. Its been a good area for us to learn from creative teams that have real unique capability."As for mobile, the platform is part of the reason why Microsoft wanted to snatch up Activision Blizzard and Candy Crush creator King. In October, Tencent-owned TiMi released a mobile version of Xbox's Age of Empires series, and he said Xbox "definitely wants to be in that [mobile] market. When we can find teams, technology, and capability that add to what were trying to do in games, absolutely we will keep our heads up.""To reach new players, we need to be creative and adaptive of new business models, new devices, new ways of access," he continued. "Were not going to grow the market with $1,000 consoles.According to Spencer, Xbox's previously reported mobile store has been pushed back as the company continues researching the market. Despite recent progress that's allowed for third-party app stores to exist on iOS and Android platforms, he said Microsoft is still exploring how to get the eventual Xbox mobile store on phones. A web store would get around the problem, but he acknowledged to Bloomberg that Microsoft "still has to have a way to find the store. I think just hoping 'if we build it, they will come' doesn't work."Spencer's full thoughts on expanding the Xbox platform, and its potential hardware future, can be read here.Read more about:[Company] XboxAbout the AuthorJustin CarterContributing Editor, GameDeveloper.comA Kansas City, MO native, Justin Carter has written for numerous sites including IGN, Polygon, and SyFy Wire. In addition to Game Developer, his writing can be found at io9 over on Gizmodo. Don't ask him about how much gum he's had, because the answer will be more than he's willing to admit.See more from Justin CarterDaily news, dev blogs, and stories from Game Developer straight to your inboxStay UpdatedYou May Also Like

Ubisoft has found a new chief people officer, following the departure of previous CPO Anika Grant after a brief two-year stint. On November 20, "seasoned HR executive" Sbastien Froidefond will take over the role.With over 30 years of experience, his duties will include handling the studio's talent and leadership development and global recruiting, along with "ensuring that HR contributes significantly to the company's overall strategic objectives." As CPO, Froidefond will be a part of Ubisoft's executive committee and report directly to Ccile Russeil, the EVP who runs its communications, HR, and legal departments.In the announcement, Ubisoft notes his prior work "demonstrated his versatility and impact across diverse industrieswhether in sectors with long, innovation-driven investment cycles, or in fast-paced environments requiring operational agility." Previously, Froidefond worked in dairy and healthcare at companies such as Lactalis Americas and Sanofi, where he managed their respective HR departments across multiple countries.His hiring marks the newest executive shakeup for Ubisoft this year. In April, Russeil was appointed to her current position after previously operating as its chief legal advisor since 1995. That same month, publishing VP Brenda Panagrossi was made general manager for its larger United States team.Froidefond's predecessor, Grant, was brought on shortly after reports surfaced of harassment and other inappropriate conduct from high-level Ubisoft workers. Ubisoft then promised to improve its culture, and she was said to be "instrumental" in that process.He joins Ubisoft at an equally tumultuous time: not only is the company facing a potential buyout between Tencent and it longtime owners, the Guillemot family, workers at its France and Milan offices have recently gone on strike over its return-to-office mandates. The studio has also been twice-sued in recent times, and is in mid-development on its next big release, Assassin's Creed Shadows for February 2025.

Apple launches Final Cut Pro 11 with even more AI featuresApple launches Final Cut Pro 11 with even more AI features / The update adds automatic masking, autogenerated captions, and spatial video editing.By Vjeran Pavic, a supervising video producer who also reviews gadgets and photography gear. He has over a decade of experience as a reporter and photojournalist, previously at Recode. Nov 13, 2024, 11:00 PM UTCShare this storyIts been 25 years since the first Final Cut Pro was announced.More than a decade after the launch of Final Cut Pro X, Apples video editing software is taking a step forward. The appis now being updated to Final Cut Pro 11, after dropping the number in its name for the past few years. The update includes new AI masking tools, the ability to generate captions directly in your timeline, spatial video editing features, and a set of workflow improvements.The new version is free for existing users and a $299 one-time purchase for new users. Final Cut Pro for iPad and Final Cut Camera are also getting some updates today, too.Ive spent the last week testing out these new features, and many of them are great improvements. Ive been particularly impressed by the speed and accuracy of one new feature coming to the desktop: Magnetic Mask. With one click, you can easily isolate a subject, like a person, from the background and apply different color adjustments to that part of the footage.I tested Magnetic Mask in various scenarios, like static talking head videos and fast-moving snowboarding footage. In each scenario, Final Cut Pro did a very good job of isolating the subjects. But dont expect a pixel-perfect mask each time. I still had to jump in and do a few smaller adjustments to help it out. You can either manually fine-tune your mask with a brush or add or remove tracking points and let Final Cut Pro analyze the footage. One thing that was impressive is that it automatically detected my flapping backpack straps.I was impressed by the speed of the whole process. Granted, these were fairly short clips (about 45 seconds each), but each mask took less than a minute on my four-year-old 10-core M1 Pro MacBook Pro a lot less time than the tedious and exhausting process of manually rotoscoping in After Effects.I did notice that analysis slowed down significantly once I started screen recording my process. This feature will work on Intel-based Macs as well. I am an avid user of Adobes Premiere Pro, but features like these always make me want to give Final Cut Pro another shot. I may not be left behind for long, though: Adobe announced a similar feature for Premiere earlier this year. DaVinci Resolve also already has a similar feature called Magic Mask.The next new highlight is the ability to autogenerate captions in your timeline. Final Cut Pro does this using an Apple-trained language mode, and the whole process takes place locally on-device without sending information to the cloud. The process is fast but not always accurate and often misspells common words. It fumbled at proper nouns like The Verge and even more common nouns like machine, where it would just write macine instead. Those are just a few of many examples. Theres also no way to stylize your captions if you were hoping to add them to your TikToks. For that, youll need to look into some third-party plug-ins. 1/4Final Cut Pro would frequently make spelling errors in the generated captions.1/4Final Cut Pro would frequently make spelling errors in the generated captions.It is a good update, but I wish Apple went a step further and added text-based editing, which lets you edit videos solely by using text instead of on the timeline itself. Text-based editing in Premiere has helped me immensely when working on longer documentaries or sit-down interviews, and I wish it were possible in Final Cut Pro 11.Other changes include the ability to edit spatial videos for the Vision Pro and some new keyboard shortcuts. My favorite is Option + Arrow Up / Arrow Down to move clips between layers. Its the little things!Final Cut Pro for iPad is also getting a few new updates. The AI-enabled enhance light and color tool that was initially released in Final Cut Pro 10.8 for the Mac is making its way to the iPad app. It is the fastest way to quickly improve the color, contrast, and overall tonality of your footage.The AI-enhanced light and color tool originally came out for Final Cut Pro for Mac but has made its way to the iPad version in this update.In a few tests that I ran, the tool did a really good job of getting me started on my coloring process. It cleans up the overall exposure nicely and adds very subtle stylistic color choices. For far less subtle color grades, Apple is expanding the number of presets available in the app. In addition to those presets, Apple is also adding new modular transitions and new songs to its soundtrack library.If youre using the Apple Pencil Pro to edit, youll finally be able to unlock all those new brushes that were released alongside the M4 iPad Pro and features like tilt recognition and take advantage of the haptic feedback. Haptic feedback works particularly well on the iPad, and Im enjoying it more than I thought I would. It really makes the editing process feel a lot more tactile. In fact, I wish more gestures had some haptic feedback.Lastly, there are some minor but significant workflow improvements. You can now resize the height of your clips in your timeline by using the pinch gesture, you can edit in 120fps timelines, and the picture-in-picture mode is dynamic. Apple also mentioned there are new keyboard shortcuts, but Ive only found one: Render Entire Timeline.Im glad to see more frequent updates coming to the iPad version of Final Cut Pro, but there are still features that are desperately needed to really make this app worth the $4.99 a month subscription plan. At the top of my wish list are things like custom LUTs, better file management, and some of the other AI-powered features that have already made their way to the desktop version. Since my initial review, Ive mostly started using DaVinci Resolve on the iPad, which continues to impress me with how similar it is to its desktop equivalent.The trifecta of updates ends with the Final Cut Camera, which can now film HEVC files in Apple Log no need to stick with storage-hungry ProRes files anymore. HEVC Log capture will work for both standalone capture or as part of the Live Multicam session. Final Cut Camera will also include LUT previews during recording, meaning you can monitor your exposure and color while filming in Apple Log.Final Cut Camera can shoot in 120fps in Apple Log and has a new leveler to help you frame up your shots.And in order to make sure your framing is correct and aligned, Apple is adding a new level indicator to the app. The new leveler includes tilt and roll indicators and crosshairs for your top-down and bottom-up shots, too.The introduction of new AI features and workflow improvements mark significant steps forward for content creators, but they dont address some of the communitys requests to fully compete with the likes of DaVinci and Adobe. Id still like to see text-based editing, more robust coloring options, and custom captions. It will be interesting to see if these new updates convert any new users. Magnetic Mask alone could be enough of a reason to switch.Most PopularMost Popular

Apple today announced some minor updates to Logic Pro for both the Mac and the iPad, including the ability to search for plug-ins and sources and the addition of more analog-simulating sounds. In Logic Pro for Mac 11.1 and Logic Pro for iPad 2.1, you can now reorder channel strips and plug-ins in the mixer and plug-in windows to make it easier to organize the layout of an audio mix. As for the new sounds, Apple added a library of analog synthesizer samples called Modular Melodies, akin to the Modular Rhythms pack already found in Logic. A more exciting sonic addition is the new Quantec Room Simulator (QRS) plug-in, which emulates the vintage digital reverb hardware of the same name, found in professional recording studios all over the world. Apple has acquired the technology for the classic QRS model and the later YardStick models to integrate into this software. I wish the QRS plug-in looked like the real life reverb unit Image: AppleSpecific to Logic Pro for Mac, you are now able to share a song to the Macs Voice Memos app which may be a great feature for when Voice Memos gets that multitrack option on the iPhone in iOS 18.2 Added to the iPad version of Logic Pro is the ability to add your own local third party sample folders to the browser window, to make it easier to bring external audio files into tracks and sampler plug-ins.These upgrades are small for current Logic users, but they do overall make the digital audio workplace easier to use and adds to the plethora of useful tools for no additional cost. Users will have access to upgrade to Logic Pro for Mac 11.1 and Logic Pro for iPad 2.1 today.

A mere 5 weeks after the launch of Unreal Engine 5.5 preview and Seattle Unrealfest Epic Games have released Unreal Engine 5.5. Jam packed with new features including vastly improved animation tools, the marquee feature of Unreal Engine 5.5 is the new experimental MegaLights, which has been called the nanite for lighting.MegaLights is a whole new direct lighting path in Unreal Engine 5 enabling artists to place orders of magnitudes more dynamic and shadowed area lights than they could ever before.MegaLights is designed to support current generation consoles and leverages ray tracing to enable realistic soft shadows from various types of area lights.MegaLights not only reduces the cost of dynamic shadowing, it also reduces the cost of unshadowed light evaluation, making it possible to use expensive light sources, such as textured area lights, on consoles.MegaLights is considered an experimental feature and should not be used in a production project.Other features of Unreal Engine 5.5 include:Improved animation toolingModular control rig is now betaMetaHuman animator now has the ability to generation facial animations from just an audio performanceMutable character customization(beta)Lumen can now run at 60hz on supported platformsPath Tracer is now production readySubstrate material system now betaMobile rendering improvements and previewerMuch, much moreKey LinksUnreal Engine 5.5 BlogUnreal Engine 5.5 Full Release NotesYou can learn more about the Unreal Engine 5.5 release in the video below. Also note, this is the first version of Unreal Engine under the new pricing structure and you will need to approve of an updated EULA before you can update to UE 5.5.

The demand for fine-tuning LLMs to incorporate new information and refresh existing knowledge is growing. While companies like OpenAI and Google offer fine-tuning APIs that allow LLM customization, their effectiveness for knowledge updating remains to be determined. LLMs used in fields like software and medicine need current, domain-specific informationsoftware developers need models updated with the latest code, while healthcare requires adherence to recent guidelines. Although fine-tuning services offer a way to adapt proprietary, closed-source models, they lack transparency regarding methods, and limited hyperparameter options may restrict knowledge infusion. No standardized benchmarks exist to evaluate these fine-tuning capabilities.Current methods to alter LLM behavior include SFT, RLHF, and continued pre-training. However, the effectiveness of these approaches for knowledge infusion is still being determined. Retrieval-augmented generation (RAG) introduces knowledge in prompts, though models often ignore conflicting information, causing inaccuracies. Past research has explored knowledge injection in open-source LLMs using adapters or shallow layer fine-tuning, but more understanding is needed around fine-tuning larger commercial models. Prior studies have fine-tuned models for classification and summarization, yet this work uniquely focuses on knowledge infusion and compares multiple fine-tuning APIs on a shared dataset.Stanford University researchers have developed FineTuneBench, a comprehensive framework and dataset to evaluate how effectively commercial fine-tuning APIs allow LLMs to incorporate new and updated knowledge. Testing five advanced LLMs, including GPT-4o and Gemini 1.5 Pro, in two scenariosintroducing new information (e.g., recent news) and updating existing knowledge (e.g., medical guidelines)the study found limited success across models. The models averaged only 37% accuracy for learning new information and 19% for updating knowledge. Among them, GPT-4o mini performed best, while Gemini models showed minimal capacity for knowledge updates, underscoring limitations in current fine-tuning services for reliable knowledge adaptation.To evaluate how well fine-tuning can enable models to learn new information, researchers created two unique datasets: a Latest News Dataset and a Fictional People Dataset, ensuring none of the data existed in the models training sets. The Latest News Dataset, generated from September 2024 Associated Press articles, was crafted into 277 question-answer pairs, which were further rephrased to test model robustness. The Fictional People Dataset included profile facts about fictional characters, producing direct and derived questions for knowledge testing. Models were trained on both datasets using various methods, such as masking answers in the prompt. Different configurations and epochs were explored to optimize performance.Fine-tuning OpenAI models shows high memorization but limited generalization for new knowledge tasks. While models like GPT-4o-mini excel in recalling trained QA pairs, they struggle with rephrased questions, especially in the Fictional People dataset, where responses to secondary or comparative questions remain weak. Updating knowledge is harder, notably in coding tasks, due to challenges in altering pre-existing information. Gemini models underperform across tasks and need help memorizing or generalizing effectively. Training methods like word masking and prompt completions also fail to enhance generalization, suggesting that standard training paradigms may not adequately improve adaptability.The study presents FineTuneBench, a dataset collection testing fine-tuned LLMs capacity to acquire knowledge in the news, fictional people, medical guidelines, and code libraries. Despite fine-tuning, models showed limited knowledge adaptation, with GPT-4o-mini outperforming others and Gemini underperforming. Relying on LLM fine-tuning remains challenging, as current methods and parameters from OpenAI and Google are limited. RAG approaches are also suboptimal due to cost and scaling issues. Limitations include testing only two LLM providers and using mostly default fine-tuning parameters. Future work will explore how question complexity impacts model generalization.Check out the Paper and GitHub Page. All credit for this research goes to the researchers of this project. Also,dont forget to follow us onTwitter and join ourTelegram Channel andLinkedIn Group. If you like our work, you will love ournewsletter.. Dont Forget to join our55k+ ML SubReddit. Sana Hassan+ postsSana Hassan, a consulting intern at Marktechpost and dual-degree student at IIT Madras, is passionate about applying technology and AI to address real-world challenges. With a keen interest in solving practical problems, he brings a fresh perspective to the intersection of AI and real-life solutions. Upcoming Live LinkedIn event, 'One Platform, Multimodal Possibilities,' where Encord CEO Eric Landau and Head of Product Engineering, Justin Sharps will talk how they are reinventing data development process to help teams build game-changing multimodal AI models, fast

10 Effective Strategies to Lower LLM Inference Costs 0 like November 13, 2024Share this postAuthor(s): Isuru Lakshan Ekanayaka Originally published on Towards AI. This member-only story is on us. Upgrade to access all of Medium.image sourceLarge Language Models (LLMs) like GPT-4 have transformed industries by enabling advanced natural language processing, content generation, and more. However, deploying these powerful models at scale presents significant challenges, particularly regarding inference costs. High operational expenses can hinder scalability, profitability, and sustainability, making it crucial to optimize LLM inference processes. This article explores ten proven strategies to reduce LLM inference costs, ensuring that AI applications remain efficient, scalable, and economically viable.Optimizing LLM inference costs isnt just a financial consideration it directly impacts several critical aspects of AI deployment:Scalability: Cost-efficient inference allows organizations to scale AI applications without prohibitive expenses, facilitating broader deployment across various use cases and markets.Profitability: Reducing operational costs directly enhances the bottom line, making AI solutions more financially viable and attractive to stakeholders.Sustainability: Optimizing inference processes can lead to reduced energy consumption, contributing to environmentally sustainable practices.Key Insight: Optimizing LLM costs is essential for scaling AI effectively and sustainably, ensuring organizations can deploy powerful AI solutions without compromising economic or environmental factors.With these considerations in mind, lets delve into ten strategies to significantly lower LLM inference costs.image sourceQuantization is a technique in machine learning Read the full blog for free on Medium.Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming asponsor. Published via Towards AITowards AI - Medium Share this post

Google Launches New AI Learning Companion Tool 0 like November 13, 2024Share this postLast Updated on November 13, 2024 by Editorial TeamAuthor(s): Get The Gist Originally published on Towards AI. Plus: Amazon Develops Its Own Custom AI ChipsThis member-only story is on us. Upgrade to access all of Medium.Welcome to Get The Gist, where every weekday we share an easy-to-read summary of the latest and greatest developments in AI news, innovations, and trends all delivered in under 5 minutes! In todays edition:Google Launches New AI Learning Companion ToolAmazon Develops Its Own Custom AI ChipsGoogle DeepMind Open-Sources AlphaFold 3Baidu Launches AI-Powered Smart GlassesAnd more AI news.Image by: GoogleThe Gist: Googles Learn About AI tool, powered by the LearnLM model, provides educational-style responses that go beyond typical chatbot answers, adding context, visuals, and interactive learning elements.Key Details:Unlike traditional chatbots, Learn About uses educational formatting with features like vocabulary builders and why it matters sections to deepen understanding.Responses are visual and interactive, using educational sources and offering related topics for further exploration.The tool successfully interprets even tricky questions, showing promise for refining AIs role in learning and answering common misconceptions.Image by: BloombergThe Gist: Amazon is advancing its in-house AI chips to decrease reliance on NVIDIAs powerful GPUs, leveraging its custom processors to cut costs and boost AI capabilities for cloud and AI partner applications.Key Details:Amazons Annapurna Labs, which previously developed the Graviton and Trainium processors, is now creating Read the full blog for free on Medium.Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming asponsor. Published via Towards AITowards AI - Medium Share this post