Mohan K’s Post

Study up on the “hidden” types of AI with Rudina Seseri, the founder and managing partner at Glasswing Ventures, as she joins #theCUBE’s Managing Director, Paul Gillin, at #Supercloud7. “When it comes to gen AI, if I were to speak to it in the context of the enablers, you're really talking about transformer and attention technologies and methods. Aside from those, we have an architecture called recurrent neural nets where it is a type of deep learning architecture,” shares Seseri. “With it, you can move the data bidirectionally. It has memory and is very powerful. You have sub architectures and techniques like PINNs, which stands for physics informed neural nets, where you can bring the boundaries of the physical world to the algorithms to mirror the physical world with incredible ramifications, ranging from life sciences to oil and gas. That has nothing to do with gen AI, but it's one of many examples of what else is happening in the AI world,” she adds. 📺 Gain more firsthand insights: https://lnkd.in/ggAzahGU

🤖✍️ From Scribbles to Silicon: The CNN is actually from 1989! Back then - while we were rocking mullets and neon, Yann LeCun and his team were quietly revolutionizing AI. They trained LeNet 1, a Convolutional Neural Network (CNN) that could actually read handwritten numbers! Picture this: You scrawl some digits on paper, and a computer goes, "Aha! That's a 7!" Mind-blowing stuff for the time. Fast forward to today, and CNNs are the secret sauce behind: • Your phone recognizing your face • Self-driving cars not crashing into trees • Instagram knowing it's a cat pic (again) From ResNet to MobileNet, CNN's digital descendants are everywhere. If it involves predicting something from an image or video, chances are there's a CNN working its magic behind the scenes. So, next time you effortlessly unlock your phone with your face, give a little nod to LeNet 1 – the OG of machine vision. Did you know the roots of AI research actually date back to as far as the 1930s? Cybernetics rings a bell? Let's geek out in the comments! 🤓 👉 Follow me-> Andreas Fraunberger 🔔 Ring the bell button on top of my profile 👉 For more about the latest in AI and augmented computing for business, follow the monthly AI XR Executive Brief: https://lnkd.in/dermJAmp #AIHistory #MachineLearning #ComputerVision #TechRevolution

In the last few days, we've seen a fresh breakthrough that has the tech world buzzing like a supercharged server farm! 🖥️⚡ One recent article dives into how a novel AI model is pushing boundaries, showing us that the sky is NOT the limit—it’s just the beginning. 🚀🌌 Key takeaway? This model is not just "thinking" outside the box; it’s redefining the box altogether. 📦🔄 It employs a blend of neural networks and quantum algorithms (yes, you heard that right—quantum!) to optimize problem-solving capabilities at unprecedented speeds. This could mean faster data processing and more efficient resource management in industries from healthcare to finance. In simpler terms, this innovation is like upgrading your brain from a regular bicycle to a sleek, AI-powered supersonic jet. 🧠➡️✈️ The implications are massive: expect smarter AI companions, improved automation, and a future that's starting to look a whole lot like science fiction. 📚🔮 Let's buckle up for this exciting journey where AI doesn't just support our future—it *is* our future. #AI #Innovation #FutureTech #QuantumComputing #AcceleratingExcellence 🚀🤓 P.S. Don't worry, Skynet's not happening yet... I hope! 😉

🌟 𝗔𝗿𝗲 𝗬𝗼𝘂 𝗥𝗲𝗮𝗱𝘆 𝘁𝗼 𝗟𝗼𝗼𝗸 𝗕𝗲𝘆𝗼𝗻𝗱 𝘁𝗵𝗲 𝗛𝗼𝗿𝗶𝘇𝗼𝗻 𝗼𝗳 𝗔𝗜? 🌟 While many are still navigating today’s #AI breakthroughs, at #OmphalosFund we’re asking: 𝗪𝗵𝗮𝘁’𝘀 𝗻𝗲𝘅𝘁? 🚀 Starting this Friday, we launch on our blog #BehindTheCloud our brand-new 8-week series: "𝗕𝗲𝘆𝗼𝗻𝗱 𝘁𝗵𝗲 𝗙𝗿𝗼𝗻𝘁𝗶𝗲𝗿: 𝗪𝗵𝗮𝘁’𝘀 𝗡𝗲𝘅𝘁 𝗳𝗼𝗿 𝗔𝗜 𝗦𝘆𝘀𝘁𝗲𝗺𝘀 𝗶𝗻 𝗔𝘀𝘀𝗲𝘁 𝗠𝗮𝗻𝗮𝗴𝗲𝗺𝗲𝗻𝘁?" From quantum computing to advanced multi-agent systems, from RAG pipelines to the future of neural networks, this series dives deep into cutting-edge AI trends that will reshape how we approach #data, #transparency, and decision-making in #AssetManagement. 𝗛𝗲𝗿𝗲’𝘀 𝗮 𝗴𝗹𝗶𝗺𝗽𝘀𝗲 𝗼𝗳 𝘄𝗵𝗮𝘁’𝘀 𝗰𝗼𝗺𝗶𝗻𝗴: 🔹 How next-gen AI models are scaling to become bigger, smarter, and more specialized. 🔹 Why multi-agent setups hold the key to collaborative AI intelligence. 🔹 The role of quantum breakthroughs in unlocking unimaginable computational power. 🔹 The evolution of AI transparency and ethics—essential for trust in the financial sector. 💡 𝗪𝗵𝘆 𝗻𝗼𝘄? The #future waits for no one. While others focus on today’s tools, we’re exploring tomorrow’s solutions — the technology trends that will define the next decade of AI in finance. ✨ Be part of the conversation. Follow our #blog (https://lnkd.in/egtH_4u7), join us every Friday, and share your thoughts. 𝙒𝙝𝙖𝙩 𝙩𝙧𝙚𝙣𝙙𝙨 𝙙𝙤 𝙮𝙤𝙪 𝙩𝙝𝙞𝙣𝙠 𝙬𝙞𝙡𝙡 𝙨𝙝𝙖𝙥𝙚 𝙩𝙝𝙚 𝙣𝙚𝙭𝙩 𝙛𝙧𝙤𝙣𝙩𝙞𝙚𝙧?  Drop your ideas in the comments! 👇 #AI #OmphalosFund #Innovation #FinTech #AIFunds #FutureofAI #QuantumComputing #RAGPipelines #EthicsInAI Borno Janekovic Pawel Skrzypek Tomasz Przeździęk Wilson “Bill” Santos Carsten Böhme

In the previous article, we explored the History of AI. Today, we’ll dive into how AI systems strive to emulate human cognition, laying the groundwork for developing intelligent machines. #AI

Our next EMTECH webinar is just around the corner! Join us on Thursday, June 27 for this edition, AI Chip Market with Anand Joshi, Market Research Director for AI at TechInsights. Anand will share his perspective on the evolution of the AI Chip market. He will discuss the success and failure stories leading up to today’s market and explore what the future will look like with the emergence of generative AI. You will have the opportunity to hear Anand dive into the transformation of the market since the inception of AI chips five years ago and assess their performance. This session will also feature in-depth analysis of both successful and unsuccessful ventures, aiming to uncover the factors that contributed to their outcomes. Additionally, the webinar will provide a comprehensive overview of the prominent AI and neural network trends, as well as the evolving need for hardware acceleration and much more. Learn more and register today: https://hubs.la/Q02zWtdV0. #GSA #WhereLeadersMeet #30yearsofgsa #semiconductor #semiconductorindustry #technology #ai #globalcollaboration #software #systems #solutions #emergingtechnologies

What’s the secret sauce behind smarter data-based decisions? Context. Neural networks are cool, but let’s be honest—they are a black box. You put data in, but good luck figuring out how they got to that decision. That’s where knowledge graphs step in, adding one magic ingredient: context. Here’s why that matters: 🤝 Seeing the big picture Knowledge graphs don’t just process data; they reveal how everything connects. Think of it as creating a clear map of a previously uncharted territory. Now, you can see how products, customers, and transactions are linked—making your data infinitely easier to navigate. 😌 Transparency at its best While neural networks give us results, they often leave us scratching our heads. Knowledge graphs? They show you exactly how a conclusion was reached. Clear pathways = smarter, more informed decisions. 🔗 Effortless integration No more wrestling with retraining models or constant adjustments. Knowledge graphs slot right into your existing data ecosystem, seamlessly connecting your sources and keeping everything in sync. Curious for more? Catch Inna Tokarev Sela's interview with SiliconANGLE & theCUBE from last month for more insights. #SemanticAI #AIGovernance #DataGovernance #DataFabric #GenAI

Exploring the Dynamics of Generative Adversarial Networks (GANs) In the evolving landscape of artificial intelligence, Generative Adversarial Networks, or GANs, stand out for their unique architecture and capabilities. This innovative AI technique involves two neural networks—the Generator and the Discriminator—locked in a fascinating digital dance. -The Generator's role is akin to an artist, crafting new, synthetic outputs such as images or texts from scratch. On the other side, the Discriminator acts as the critic, tasked with differentiating the artificial creations from real, authentic works. -The true power of GANs lies in their iterative training process. With each round, the Generator strives to enhance its artistry, fooling the Discriminator with ever-more convincing fakes. Concurrently, the Discriminator sharpens its ability to detect nuances, pushing the Generator to new heights. This dynamic rivalry drives both networks towards perfection, leading to the creation of remarkably realistic and high-quality synthetic content. -Whether it's crafting new artworks, simulating virtual environments, or developing new drugs, GANs are paving the way for a future where the lines between artificial and real increasingly blur. (Dive into the visual metaphor of this fascinating technology in the image attached, illustrating the interplay between creation and critique within GANs.) #ArtificialIntelligence #MachineLearning #GenerativeAI #TechnologyInnovation #DigitalTransformation

The Journey of Data: Exploring Forward Propagation in ANNs Forward propagation in an Artificial Neural Network (ANN) is the process by which input data passes through the network layers to generate the output. It involves computing the activations of each neuron layer by layer, starting from the input layer and moving through hidden layers to the output layer Input Layer: The network starts by receiving input data, like a set of features. Weights and Biases: Each connection between neurons has a weight, and every neuron has a bias that influences the output. Initially, these are set randomly. Activation Calculation: Each neuron calculates a value by multiplying the input data by its weights and adding the bias. This value determines the neuron's output. Activation Function: The calculated value is passed through an activation function (like ReLU, Sigmoid, or Tanh), introducing non-linearity, which helps the network capture complex relationships in the data. Output Layer: After the data passes through the hidden layers, it reaches the output layer, where the final prediction is made. For a binary classification task, this might be a probability score between 0 and 1, representing the likelihood of an outcome Forward propagation is a key step during both training and inference, but during training, it is followed by backpropagation, where the network updates the weights and biases to minimize errors based on the output. Tomorrow let's look into back Propagation #ai #ml #machinelearning #datascience #ffnn #ann #forwardpropogation #deeplearning #dl #post #every #day #blog

🚀 Fast-RCNN: Making Object Detection Faster & Smarter! 🚀 Object detection models are game-changers in computer vision 🌍. But they’re not all the same! One model that pushed the boundaries significantly is Fast-RCNN (Region-based Convolutional Neural Networks). Here’s why it’s worth knowing: 🌟 What is Fast-RCNN? Fast-RCNN is an evolution of the original R-CNN model, designed to detect and classify multiple objects within images 📸 – but way faster! 🚅 ⚡ Why Fast-RCNN Stands Out: 1. Single-Stage Training 🏋️♂️ - Unlike R-CNN, which uses multiple stages, Fast-RCNN optimizes the entire process in one single stage! Less complexity, more efficiency! 💯 2. End-to-End Learning 🎯 - With Fast-RCNN, the model learns to detect and classify objects all at once, making it an end-to-end solution for object detection 💡. It uses the Softmax function for classification and a bounding box regressor for precise object location. 📍 3. ROI Pooling Magic ✨ - Fast-RCNN introduced the Region of Interest (ROI) Pooling layer, transforming different regions of interest into a fixed size 📏. This speeds up the process without compromising accuracy – it’s efficient and powerful! 💥 🔥 Key Benefits: - Faster Processing ⚙️ – Cuts down computational time dramatically! - Less Memory Intensive 💾 – Requires less storage and fewer resources. - Higher Accuracy 🎯 – Detects objects more accurately than its predecessor, R-CNN. 📈 Impact on Real-World Applications: Whether it’s self-driving cars 🚗, facial recognition 🧑💻, or medical imaging, Fast-RCNN plays a role in making these technologies faster and more reliable! 💪 Ready to dive into the future of computer vision? Fast-RCNN set the foundation for advanced models like Faster-RCNN and YOLO that continue to shape today’s AI advancements! 🚀 #ComputerVision #MachineLearning #FastRCNN #AI #ObjectDetection #DeepLearning #Innovation #ml #ML