John Wiseman

Happy 2025 to all! CES announcement from NVIDIA for game developers including Neural Rendering, Neural Face, and new real time diffusion based Audio2Face. Check it out. https://lnkd.in/gkndykPE

46

Well worth going through the paper or slides, super well explained.

18

🎓 Celebrating my graduation from the prestigious UC Berkeley Chief Technology Officer Program! 🎓 I want to celebrate a special mentor in my life.  Many of us have dreams, and some of us have goals.  Few of us have knowledgeable supporters, in our lives, that encourage us to move forward on our dreams!  Winston She, thank you for being that person for me.  Your encouragement saved me many unnecessary years of delay. This incredible journey, which began in January 2023 and concluded in June 2024, has been transformative, equipping our cohort with advanced leadership and AI skills to excel in today’s rapidly evolving technological landscape. Congratulations to all my fellow graduates! The bonds we've formed over the past year and a half will last a lifetime. The program provided a comprehensive dive into: 📌 Digital Transformation: From data to actionable insights. 📌 Machine Learning and Neural Networks: Practical applications in various domains. 📌 Sequential Decision Making and Reinforcement Learning: Impact on decision-making processes. 📌 Computer Vision: Enhancing understanding through case studies. 📌 Enabling Technologies with Data: Core principles for advancements. 📌 Business Model Canvas and Jobs to Be Done: Analyzing models to drive innovation. 📌 Leading Digital Strategy: Digital strategies, organizational change, social networks, future of work, enterprise agile, decision making, negotiating, innovation portfolio management, and agile mindset. 📌 Digital Architecture and Operations: Product architecture and go-to-market strategies. 📌 Business Analytics and AI: AI-centric operating models. 📌 Descriptive Analytics: Transforming and visualizing data. 📌 Predictive Analytics: Logistic regression, decision trees, random forests, and neural networks. 📌 Unsupervised Learning: Clustering and principal component analysis. 📌 Reinforcement Learning: Q-learning and A/B testing. 📌 Experimentation: Designing and implementing A/B tests. 📌 Future of Big Data and AI: Ethical challenges and data management. 📌 Tech Futures: Emerging technologies with Gartner’s Hype Cycle and S-Curve. 📌 AI and Data Science: Creating a data-driven culture. 📌 Compute Power: Exploring edge, neuromorphic, and quantum computing. 📌 Internet of Everything (IoE): Applications in AR, VR, and multi-sided platforms. 📌 Robotics and Automation: Human-machine productivity strategies. 📌 Cybersecurity: Threats and risk reduction strategies. 📌 Fintech and Blockchain: Blockchain disruptions in financial services. 📌 Portfolio Strategy: Managing innovation initiatives. 📌 Capstone Projects: Applying learned concepts to real-world challenges. #ClassOf2024 #Graduation #UCBerkeleyCTOProgram #Leadership #CTO #AI #DigitalTransformation #Innovation #TechLeadership #CareerGrowth #LifelongLearning #BerkeleyExecEd #BerkeleyHaas #Leader #ChiefTechnologyOfficer #ExecutiveEducation #BerkeleyCTO

104

31 Comments

NVIDIA CEO Jensen Huang Keynote at CES 2025 "The ChatGPT moment for general robotics is just around the corner" - Jensen Huang Jensen talked about a multitude of innovations by NVIDIA at the CES 2025 today. It has touched upon most industrial fronts ranging from automobiles to AI and beyond! #AI #Nvidia #JensenHuang #Technology #CES2025 #Robotics

5

2 Comments

When I met Surjyanil (Tony) Chowdhury and Stephen Barron at NVIDIA GTC in San Jose this year, what they showed me in CesiumJS was amazing. They were bringing fire and weather data from the National Interagency Fire Center (NIFC), Geostationary Operational Environmental Satellites (GOES), NOAA, and ADS-B flight trackers and combining them in CesiumJS. All on top of Cesium World Terrain and Bing Maps Aerial imagery streamed as 3D Tiles with Cesium ion. Their idea is to give wildfire incident commanders and staff an accurate 3D map to visualize and analyze active wildland fires, track flights of the air assets doing detection, monitoring, and suppression, and then incorporate critical weather information, such as temperature, humidity, and wind. It makes sense they're using Cesium software - CesiumJS is built for interactive web apps and streaming massive amounts of data. It's lightweight and performant, which is vital for users receiving rapid updates in the field. A couple months after GTC, Steve demo'd it to the Cesium user community at Cesium's office in Philly during our inaugural 3D Technical Exchange Meeting (3D TEM). Just beautiful. Their next ambitious goal is to detect near-real-time fire perimeters from imagery using Lockheed's AI tools, and reflect them on top of Cesium World Terrain every 15 to 20 minutes. The approach is to refine the images prior to generating the near-real-time fire perimeters from the satellite imagery. Learn more here: https://lnkd.in/eEKdV_rN #3d #geospatial #wildfires #simulation #ai Christopher Tucker Trent Tinker Alex Paulson Klaus Konarkowski David Sracic

48

3 Comments

Exciting news in analog IC migration! Read this SemiWiki article on how AI-powered Virtuoso Studio layout tools offer an automated migration flow that streamlines schematic and layout migration to achieve up to 2X productivity gains over manual methods. Learn how Cadence’s flow saves time and reduces engineering effort. https://ow.ly/Cz3X50U8ouB #ICDesign #AnalogMigration #Cadence #AIDriven

9

At #CES2025, NVIDIA CEO Jensen Huang announced a cornucopia of product innovation. I picked six to write about - check out my SiliconANGLE & theCUBE post that looks at 1) Cosmos 2) Omniverse 3) Blueprints for developers 4) RTX50 5) Project DIGITS 6) Enabling agentic AI https://lnkd.in/g6xbm4ns

32

2 Comments

A new foreigner has arrived in town… #Molmo, an #opensource family of state-of-the-art #multimodal #AI models which outpeform top proprietary rivals including OpenAI’s GPT-4o, Anthropic’s Claude 3.5 Sonnet, and Google’s Gemini 1.5 on several third-party benchmarks. Molmo consists of four main models of different parameter sizes and capabilities: 1. Molmo-72B (72 billion parameters, or settings — the flagship model, based on based on Alibaba Cloud’s Qwen2-72B open source model) 2. Molmo-7B-D (“demo model” based on Alibaba’s Qwen2-7B model) 3. Molmo-7B-O (based on Ai2’s OLMo-7B model) 4. MolmoE-1B (based on OLMoE-1B-7B mixture-of-experts LLM, and which Ai2 says “nearly matches the performance of GPT-4V on both academic benchmarks and user preference.”) molmo.org looks really promising. It's going to be fun!🤩 #innovation #tech

11

1 Comment

NVIDIA Nemotron might not be the open weight, small size, state-of-the-art model that everyone was waiting for, but it is a groundbreaking advance in open weight models. My prediction is that by Q1 2025, we should have an open weight model rivalling current state-of-the-art models by OpenAI and Anthropic. We broadly understand how to get there.

1

https://lnkd.in/eXbCsAnD “He who controls the GPUs, controls the universe.” Access to GPUs, and in particular those made by Nvidia, the leading supplier, is vital for any company that wants to be taken seriously in #ai. Nvidia’s dominance has driven its market value to more than $2trn. 'GPUs do the computational heavy-lifting needed to train & operate large #ai models. Yet, oddly, this is not what they were designed for. The initials stand for “graphics processing unit”, because such chips were originally designed to process video-game graphics. It turned out that, fortunately for Nvidia, they could be repurposed for #ai workloads. 'Might it be better to design specialist ai chips from scratch? That is what many companies, small & large, are now doing in a bid to topple Nvidia. Dedicated ai chips promise to make building and running ai models faster, cheaper or both. Any firm that can mount a credible threat will have no shortage of customers, who dislike its lofty prices and limited supplies. 'GPUs contain several thousand processing engines, or “cores”, which let them run thousands of versions of the same simple task at the same time. Running #ai models similarly involves running lots of copies of the same task in parallel. Figuring out how to rewrite ai code to run on GPUs was one of the factors that triggered the current ai boom. 'Yet GPUs have their limitations, particularly when it comes to the speed with which data can be shuffled on and off them... When training very large #ai models, some GPu cores may be idle as much as half of the time. Andrew Feldman, the boss of Cerebras, a startup based in Sunnyvale, CA, likens it to the gridlock in a grocery store on the day before Thanksgiving. “Everybody’s in a queue, so there are blockages in the parking lot, there are blockages in the aisles, blockages at the checkout. That’s exactly what’s happening with a gpu.” 'Cerebras’s response is to put 900,000 cores, plus lots of memory, onto a single, enormous chip, to reduce the complexity of connecting up multiple chips & piping data between them. 'Groq, a startup from Mountain View, CA, is taking a different approach. Its ai chips, called language processing units (lpus), are optimised to run #llms particularly quickly. 'Yet another approach is that taken by MatX, also based in CA. GPUs contain features & circuitry that provide flexibility for graphics, but are not needed for llms. The firm is working on getting rid of such unnecessary cruft.

14

3 Comments

For the last two months, Ibrahim Abdalkader and I have been refactoring our TensorFlow support on the OpenMV Cam. We're super happy to finally release the new API which adds support for running multi-input and multi-output networks on the OpenMV Cam along with enabling pre/post-processing of Model input/output using NumPy. https://lnkd.in/eH4vCp-f It took some courage to think this is the way forward for ML processing on Microcontrollers on the Edge. We embraced a lot of desktop-like ideas on systems with limited RAM and compute. However, this is the right way to go as we were able to achieve performance targets, RAM efficiency, and make it much easier to run any Model you want (within system limits) simultaneously.

53

1 Comment

One additional step toward scalability: generalization across roads and cities.

8

AI is transforming transit bus fleet optimization, and I’m excited to connect at #CalTransit24 in San Jose to discuss it! Looking forward to exploring cutting-edge innovations that enhance efficiency, reduce costs, and support transit agencies in reaching zero emissions and sustainability goals. Let’s shape the future of public transportation together! #AIinTransportation #AI #ML #MachineLearning #CalTransit24 #PublicTransit #ElectricVehicles #SustainableTransport #FleetManagement #EV #Transit #Bus #Busfleet #Fleet #ArtificialIntelligence

2

Friends. Jensen Huang trash-talked quantum. That’s because NVIDIA only understands 3d modeling. Airforce and CDAO are moving toward quantum (4d+) modeling. Keep an eye out for tech that is more predictive, closer to AGI, with military tech.

8

The future is undeniable. Sensors are the new internet users. wireless is the only path forward. combine AI with the wireless and you can do things you could not do before: Look at this picture below:

37

2 Comments

Great session with Eric Schmidt on the "near" future of AI. From the next steps of AI becoming a ubiquitous utility, evolving from an "assistant" to a polymath "partner", the black box (non)issue, etc... Take some time to listen to one of the OG's of the last 50 years of computer science, the internet and oh yes...Google https://lnkd.in/e3bKw_yq

2