Joe Brown的动态

They say the algorithm is always watching ??. And it is. Every scroll, click, like, and share—it’s all feeding the machine. But here’s the thing... What if you were the one pulling the strings? What if you could crack the code, flip the narrative, and make that sneaky algorithm work for you? I’m Joe from Digital Junkies—your local dealer in all things digital. I’m not here to sell you anything (yet). Just to ask: Is your digital presence winning the game, or just playing along? Check out OUR WORK (yes, caps—because it’s that good): Or book a quick 15-min call to chat about how we can make some chaos work in your favour.

Blink and you might miss it. And by that I mean the constantly changing algorithms! If you've noticed that a particular type of content isn't performing as well as it used to (for instance, a Reel), it is likely because of a recent algorithm change. That's why it becomes crucial to remain omnipresent and to adjust accordingly.

Hi, I created a video about uninformed and informed search strategies. I also discuss the greedy best-first search, heuristic functions, and Manhattan distance. After that, I explain the A* search algorithm for finding optimal solutions. Finally, I cover the minimax algorithm with a reference to the Tic-Tac-Toe game. So, Check it out here: https://lnkd.in/emm3WXtz #SearchAlgorithms #AIandMachineLearning #TicTacToe

Temporary patches rarely work. If your attribution model is broken, maybe it's time to consider a different method? Some recommend using multipliers from experiments (true incrementality tests) to fix the results of attribution models. Treat this tactic with caution. Attribution models suffer from many issues and can be very misleading (including MTA). Relying on a broken model that may point in the wrong direction and then simply adjusting the magnitude of (wrong) ad effects is still damaging and can waste lots of media dollars. If results among methods don't agree, then it's a good idea to examine why this is the case. Does my model still work? Don't get me wrong. We generally want to use experiments to calibrate and check other methods—both MMM and attribution models. But we need to ask some hard questions and consider how to improve our models and approach in the long run. And sometimes we may need to admit that it's time to ditch an old model in favour of a new method.

well, it looks like a dystopian?cyberpunk future with greedy corporations holding us by the throat is delayed for a few years. the fresh new open-source LLM DeepSeek-R1-Zero is our savior. if you - are worried about the security of your applications,? - are worried about the privacy of your users' data,? - don't want openai and other corps to have your data,? - and want to run your local capable AI model,? ask me how. learn more here - https://lnkd.in/ePD35bSQ

Create AI-images that match your style https://lnkd.in/djAQsSS2 Crafting the perfect prompt for Midjourney or DALL-E can be challenging. With?many parameters to customize?and tweak your prompt, mastering these tools in-depth takes hundreds of hours. But here is a simple hack you can try out.

The algorithm is getting better.

People ask us why we are continuing to push on inference performance when we can already generate first tokens and overall tokens per second faster than the human eye can read. Well because we don’t believe humans will be the primary audience for the output of these models in the future. In an agentic world, where we are chaining lots of agents together, it’s imperative that the first tokens come out quickly and we produce fastest tokens per second because when we string a chain of these agents together to compose workflows, all the latency accumulates. So let’s push on latency. Let’s push on throughput!

Ready to unleash the magic of image recognition? ?? Dive into our latest project where we explore a captivating dataset featuring our furry friends - cats and dogs! ???? With 10,000 train images and a corresponding CSV file, we're embarking on an adventure with 5 unique featuring models and a showdown between 3 powerhouse machine learning algorithms: Random Forest, Naive Bayes, and Gradient Boosting. Get ready to witness the art of predicting pet presence in images come to life! ???? #ImageRecognition #MachineLearningMagic #PetPredictions https://lnkd.in/eE5hzhH8