正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Building a machine learning model isn’t just about finding the right algorithm—it’s about the right process. In Machine Learning in Production, Andrew Ng breaks down the iterative loop of model development: training, error analysis, refining hyperparameters, and improving data. Getting to a high test set accuracy is one thing,... show more

DeepLearning.AI

333,137 subscribers

45,248 次观看 • 1 年前 •via X (Twitter)

教育科学技术

Anya Rossi• Live Now

Private livecam show

5 条评论

DataInsta 的头像

DataInsta1 年前

exactly! it’s that feedback loop that turns models into masterpieces!

Rainmaker 的头像

Rainmaker2 年前

Which Machine Learning model delivers stronger trading results? Check out this free Substack post where I compare several powerful models that beat the market and show yearly returns of over 20%.

Lester Smartfield 的头像

Lester Smartfield1 年前

Technical stuff indeed! But, Andrew Ng always makes it sound feasible, right? Looking forward to diving in!

QuantumQuinn Kierra 的头像

QuantumQuinn Kierra1 年前

Models seldom salute business bosses. Sounds like a data-driven adventure awaits!

Zephyr Cristo 的头像

Zephyr Cristo1 年前

Absolutely, the process is key in ML. Ng's iterative approach highlights how continuous refinement drives model performance, much like how entrepreneurs must iterate on their business models to find success.

相关视频

Watch "Endless experimentation: Building AI models in the wild." This is a 50-minute, MIT lecture about problems when deploying models and LLMs in the real world and how to prepare to solve them. This is a great lecture for those building production machine learning.

Watch "Endless experimentation: Building AI models in the wild." This is a 50-minute, MIT lecture about problems when deploying models and LLMs in the real world and how to prepare to solve them. This is a great lecture for those building production machine learning.

Santiago

117,204 次观看 • 2 年前

VPS or visual positioning system... this is how AR glasses and robots will understand where they are in the real world and know where to go. It’s also how you can spatially annotate reality with cm level accuracy - all using a machine readable model of the world.

VPS or visual positioning system... this is how AR glasses and robots will understand where they are in the real world and know where to go. It’s also how you can spatially annotate reality with cm level accuracy - all using a machine readable model of the world.

Bilawal Sidhu

27,898 次观看 • 7 个月前

We’re tired of empty meme coins no purpose, no impact, just noise 🏰 The Emperor coin is something bigger. A movement with meaning. A bridge between the online world and real-life connection. We’re using crypto to bring people together in the real world, through powerful events, shared vision, and true progress. It’s not just about tokens. It’s about opening doors, building futures, and improving real lives. We’re here to fill the gap and be the bridge ⚔️❤️

We’re tired of empty meme coins no purpose, no impact, just noise 🏰 The Emperor coin is something bigger. A movement with meaning. A bridge between the online world and real-life connection. We’re using crypto to bring people together in the real world, through powerful events, shared vision, and true progress. It’s not just about tokens. It’s about opening doors, building futures, and improving real lives. We’re here to fill the gap and be the bridge ⚔️❤️

Emperor Coin 🏰

12,802 次观看 • 1 年前

The world around us is changing so fast. And to stay relevant, you need to constantly upskill yourself as well as accumulate capabilities. Data Analytics, Automation, and AI skill will continue to grow and evolve. On this channel, I will help you grow your knowledge of in these areas. From building your first machine learning model to deployment, building your first dashboard to becoming a pro, understanding business process automation using low code tools, data management, as well as learning from industry professionals through podcast. If you would like to learn about Microsoft Fabric, Power Platform, Data Management and Governance, Data Analytics, you are in the right place.

The world around us is changing so fast. And to stay relevant, you need to constantly upskill yourself as well as accumulate capabilities. Data Analytics, Automation, and AI skill will continue to grow and evolve. On this channel, I will help you grow your knowledge of in these areas. From building your first machine learning model to deployment, building your first dashboard to becoming a pro, understanding business process automation using low code tools, data management, as well as learning from industry professionals through podcast. If you would like to learn about Microsoft Fabric, Power Platform, Data Management and Governance, Data Analytics, you are in the right place.

TheOyinbooke

15,539 次观看 • 2 年前

Getting started with an ML project? Andrew Ng shares practical tips to set up a strong foundation—choosing a reasonable algorithm, running quick sanity checks, and focusing on data quality over chasing the latest models. These steps can save time, prevent errors, and make iteration more efficient. Streamline your workflow and build better ML systems in the Machine Learning in Production course:

Getting started with an ML project? Andrew Ng shares practical tips to set up a strong foundation—choosing a reasonable algorithm, running quick sanity checks, and focusing on data quality over chasing the latest models. These steps can save time, prevent errors, and make iteration more efficient. Streamline your workflow and build better ML systems in the Machine Learning in Production course:

DeepLearning.AI

11,307 次观看 • 1 年前

New video series: Physics Informed Machine Learning! Physics may be embedded into AI/ML in 5 stages: 1 choose what to Model 2 curate training Data 3 design an Architecture 4 craft a Loss Function, and 5 implement Optimization Algorithm to train the model

New video series: Physics Informed Machine Learning! Physics may be embedded into AI/ML in 5 stages: 1 choose what to Model 2 curate training Data 3 design an Architecture 4 craft a Loss Function, and 5 implement Optimization Algorithm to train the model

Steven Brunton

78,441 次观看 • 2 年前

Introducing the OpenRouter MCP, live model intelligence right inside your agent Your agent builds and ships, but when it comes to choosing the right model for the right job, it guesses from 6 month old training data Watch it pick, price, and test the right model:

Introducing the OpenRouter MCP, live model intelligence right inside your agent Your agent builds and ships, but when it comes to choosing the right model for the right job, it guesses from 6 month old training data Watch it pick, price, and test the right model:

OpenRouter

248,663 次观看 • 8 天前

New PNAS paper. Historical GDP per capita data is scarce, but data on the places of birth, death, and occupations of famous individuals is abundant. In this paper we estimate the historical GDP per capita of hundreds of regions in Europe and North America using a machine learning model that leveraged data on about 500k famous biographies. Our estimates more-or-less quadruple the availability of historical GDP per capita estimates for the last 700 years. So why use biographies to augment historical GDP per capita data? Biographical data contains information about people who might have contributed directly to economic growth, like James Watt, or that were attracted to wealthy places looking for patrons, like Michelangelo. So we--mainly Philipp (Philipp Koch)--used this data to construct hundreds of features describing each European region. Then, we trained a machine learning model to find the features that explained most of the variance in a cross-validation test, where we split regions multiple times into a training set and a test set. On average, the model explained about 90% of the variance in GDP per capita of the regions it had not seen during training. But we wanted to go further, and Philipp really went to town by looking at different ways to validate our estimates. We found our estimates correlate positively with historical measures of wellbeing, church building activity, urbanization, and body height. We also used these measures to reproduce the basic Atlantic trade result of Acemoglu, Johnson, and Robison and to explore the economic consequences of the famous Lisbon earthquake of 1755. But what I personally loved most about this project, other than working with Philipp Koch and V, is that it shows that we can use machine learning methods not only to explore the future, but the past. There is a bright and growing future in the use of machine learning for economic history. Hope you enjoy the paper and the data. You can find links to the paper and a data exploration tool in the first comment.

New PNAS paper. Historical GDP per capita data is scarce, but data on the places of birth, death, and occupations of famous individuals is abundant. In this paper we estimate the historical GDP per capita of hundreds of regions in Europe and North America using a machine learning model that leveraged data on about 500k famous biographies. Our estimates more-or-less quadruple the availability of historical GDP per capita estimates for the last 700 years. So why use biographies to augment historical GDP per capita data? Biographical data contains information about people who might have contributed directly to economic growth, like James Watt, or that were attracted to wealthy places looking for patrons, like Michelangelo. So we--mainly Philipp (Philipp Koch)--used this data to construct hundreds of features describing each European region. Then, we trained a machine learning model to find the features that explained most of the variance in a cross-validation test, where we split regions multiple times into a training set and a test set. On average, the model explained about 90% of the variance in GDP per capita of the regions it had not seen during training. But we wanted to go further, and Philipp really went to town by looking at different ways to validate our estimates. We found our estimates correlate positively with historical measures of wellbeing, church building activity, urbanization, and body height. We also used these measures to reproduce the basic Atlantic trade result of Acemoglu, Johnson, and Robison and to explore the economic consequences of the famous Lisbon earthquake of 1755. But what I personally loved most about this project, other than working with Philipp Koch and V, is that it shows that we can use machine learning methods not only to explore the future, but the past. There is a bright and growing future in the use of machine learning for economic history. Hope you enjoy the paper and the data. You can find links to the paper and a data exploration tool in the first comment.

César A. Hidalgo

54,330 次观看 • 1 年前

Aroon Purie lays it out clearly: India’s media may be unmatched in scale, but underneath is a crumbling business model crippled by overregulation. But the real danger? A system where survival hinges on appeasement, not independence. When business breaks, bias fills the gap. Fixing media economics isn’t just about profit, it’s about preserving trust.

Aroon Purie lays it out clearly: India’s media may be unmatched in scale, but underneath is a crumbling business model crippled by overregulation. But the real danger? A system where survival hinges on appeasement, not independence. When business breaks, bias fills the gap. Fixing media economics isn’t just about profit, it’s about preserving trust.

Anshul Saxena

47,960 次观看 • 8 个月前

The Machine That Learns The Law Behind The Data A very very interesting US Patent US10963540B2 - Physics Informed Learning Machine describes a learning system that does not begin with data alone. It begins with a physical model, usually written as a differential equation (or PDE) dx/dt = f(x,t) A normal Machine Learning model sees scattered data and tries to fit it. A physics-informed learning machine starts with a law. Then it treats the data as evidence that updates what the model believes about the physical system. For this application, I use the patent idea on NASA C-MAPSS Turbofan engine data. The machine watches multivariate telemetry from a degrading engine and infers a hidden health state that is not measured directly. From that posterior belief, it estimates the engine’s remaining useful life. In the main 3D scene, the engine lifetime is turned into a tunnel. The spiral ribbons are real sensor channels evolving over cycle-time. The glowing core is the inferred health state. The surrounding cloud is uncertainty. The orange wall ahead is the predicted failure horizon. So the big picture is: sensor evidence comes in, posterior belief tightens, and the machine moves from uncertainty toward a concrete failure prediction. The inset posteriors make that explicit. The health posterior shows where the model believes the hidden engine condition sits at the current moment, and how sharply it believes it. The RUL posterior shows the same idea for remaining life... early on it is broad, later it shifts left and narrows as the machine becomes more certain about how close failure is. This idea is not limited to engines. The same idea can apply to data centers, CPUs, GPUs, cooling systems, power grids, robotics, batteries, and any machine that produces telemetry while obeying physical constraints. In an age where machine learning runs on massive hardware infrastructure, this kind of model matters: it can turn noisy sensor streams into early warnings before expensive systems fail.

The Machine That Learns The Law Behind The Data A very very interesting US Patent US10963540B2 - Physics Informed Learning Machine describes a learning system that does not begin with data alone. It begins with a physical model, usually written as a differential equation (or PDE) dx/dt = f(x,t) A normal Machine Learning model sees scattered data and tries to fit it. A physics-informed learning machine starts with a law. Then it treats the data as evidence that updates what the model believes about the physical system. For this application, I use the patent idea on NASA C-MAPSS Turbofan engine data. The machine watches multivariate telemetry from a degrading engine and infers a hidden health state that is not measured directly. From that posterior belief, it estimates the engine’s remaining useful life. In the main 3D scene, the engine lifetime is turned into a tunnel. The spiral ribbons are real sensor channels evolving over cycle-time. The glowing core is the inferred health state. The surrounding cloud is uncertainty. The orange wall ahead is the predicted failure horizon. So the big picture is: sensor evidence comes in, posterior belief tightens, and the machine moves from uncertainty toward a concrete failure prediction. The inset posteriors make that explicit. The health posterior shows where the model believes the hidden engine condition sits at the current moment, and how sharply it believes it. The RUL posterior shows the same idea for remaining life... early on it is broad, later it shifts left and narrows as the machine becomes more certain about how close failure is. This idea is not limited to engines. The same idea can apply to data centers, CPUs, GPUs, cooling systems, power grids, robotics, batteries, and any machine that produces telemetry while obeying physical constraints. In an age where machine learning runs on massive hardware infrastructure, this kind of model matters: it can turn noisy sensor streams into early warnings before expensive systems fail.

Mathelirium

17,758 次观看 • 2 个月前

The future of robotics isn’t being built in factories, it’s being trained in simulation MasterBOT’s proprietary AI model is learning in real-time, powered by reinforcement and continuous feedback loops This is how machines learn to think $BOT

The future of robotics isn’t being built in factories, it’s being trained in simulation MasterBOT’s proprietary AI model is learning in real-time, powered by reinforcement and continuous feedback loops This is how machines learn to think $BOT

MasterBOT

51,156 次观看 • 8 个月前

notes on my reinforcement learning for rocket league side quest so far: i’m in awe that — we live in a time where i can train a reinforcement learning model that is learning to play rocket league and at the same time, test community made rocket league models. where, both processes can run in parallel & locally on my hardware. in the game window — i’m playing against a reinforcement learning ppo model that’s running locally. and in the vscode window — i’m training my own reinforcement learning model (just a dummy example script for now). also, the rocket league botting ecosystem has incredible tooling where you can test out different reinforcement learning ideas against other algorithms + it has great documentation for linux and windows. it’s honestly a beautiful hidden gem of a rabbit hole :)

notes on my reinforcement learning for rocket league side quest so far: i’m in awe that — we live in a time where i can train a reinforcement learning model that is learning to play rocket league and at the same time, test community made rocket league models. where, both processes can run in parallel & locally on my hardware. in the game window — i’m playing against a reinforcement learning ppo model that’s running locally. and in the vscode window — i’m training my own reinforcement learning model (just a dummy example script for now). also, the rocket league botting ecosystem has incredible tooling where you can test out different reinforcement learning ideas against other algorithms + it has great documentation for linux and windows. it’s honestly a beautiful hidden gem of a rabbit hole :)

naklecha

46,302 次观看 • 1 年前

The best way to get robust, high-quality robot performance is through reinforcement learning; but RL in either the real world or a traditional simulation has lots of limitations. Instead, Jiazhi Yang in RISE does RL in a compositional world model. Learn more ->

The best way to get robust, high-quality robot performance is through reinforcement learning; but RL in either the real world or a traditional simulation has lots of limitations. Instead, Jiazhi Yang in RISE does RL in a compositional world model. Learn more ->

Chris Paxton

33,766 次观看 • 21 天前

Real-world robot data is expensive and slow to collect, creating a major challenge for humanoid development. 🤖 The NVIDIA GR00T N1.6 open vision language action model is pre-trained on a diverse mix of data, including thousands of hours of Stanford Vision and Learning Lab’s BEHAVIOR simulation data, which covers long-horizon everyday manipulation tasks. This diverse training is the key to robust cross-embodiment performance and real-world adaptability. 🌍 Read the blog 🔗

Real-world robot data is expensive and slow to collect, creating a major challenge for humanoid development. 🤖 The NVIDIA GR00T N1.6 open vision language action model is pre-trained on a diverse mix of data, including thousands of hours of Stanford Vision and Learning Lab’s BEHAVIOR simulation data, which covers long-horizon everyday manipulation tasks. This diverse training is the key to robust cross-embodiment performance and real-world adaptability. 🌍 Read the blog 🔗

NVIDIA Robotics

13,429 次观看 • 5 个月前

AI is changing everything, but how does it actually work? We break down machine learning, training, inference, and what happens behind the scenes when you use AI in the latest episode of #TheBringUp, featuring a hands-on model demo. 🎥 Watch now:

AI is changing everything, but how does it actually work? We break down machine learning, training, inference, and what happens behind the scenes when you use AI in the latest episode of #TheBringUp, featuring a hands-on model demo. 🎥 Watch now:

AMD

22,825 次观看 • 6 天前

The next bottleneck in AI isn’t model performance. It’s data. From scraped internet content → real-world, consented, high-context data.

The next bottleneck in AI isn’t model performance. It’s data. From scraped internet content → real-world, consented, high-context data.

The DATA Foundation

12,457 次观看 • 18 天前

Today, we're joined by Sergey Levine, associate professor at UC Berkeley EECS and co-founder of Physical Intelligence to discuss π0 (pi-zero), a general-purpose robotic foundation model. We dig into the model architecture, which pairs a vision language model (VLM) with a diffusion-based action expert, and the model training "recipe," emphasizing the roles of pre-training and post-training with a diverse mixture of real-world data to ensure robust and intelligent robot learning. We review the data collection approach, which uses human operators and teleoperation rigs, the potential of synthetic data and reinforcement learning in enhancing robotic capabilities, and much more. We also introduce the team’s new FAST tokenizer, which opens the door to a fully Transformer-based model and significant improvements in learning and generalization. Finally, we cover the open-sourcing of π0 and future directions for their research. 🎧 / 🎥 Listen or watch the full episode on our page: 📖 CHAPTERS =============================== 00:00 - Introduction 2:14 - Physical Intelligence 3:47 - Key challenges in robotic learning 6:13 - Reinforcement learning in π0 and robotic foundation models 8:36 - π0 VLM model architecture 15:33 - π0 model recipe 18:39 - Pre-training dataset 22:47 - Post-training 24:23 - Laundry folding demo 31:32 - Scaling laws on π0 model 34:57 - FAST 40:26 - Open sourcing π0 43:37 - Other robot types 46:27 - Future directions

Today, we're joined by Sergey Levine, associate professor at UC Berkeley EECS and co-founder of Physical Intelligence to discuss π0 (pi-zero), a general-purpose robotic foundation model. We dig into the model architecture, which pairs a vision language model (VLM) with a diffusion-based action expert, and the model training "recipe," emphasizing the roles of pre-training and post-training with a diverse mixture of real-world data to ensure robust and intelligent robot learning. We review the data collection approach, which uses human operators and teleoperation rigs, the potential of synthetic data and reinforcement learning in enhancing robotic capabilities, and much more. We also introduce the team’s new FAST tokenizer, which opens the door to a fully Transformer-based model and significant improvements in learning and generalization. Finally, we cover the open-sourcing of π0 and future directions for their research. 🎧 / 🎥 Listen or watch the full episode on our page: 📖 CHAPTERS =============================== 00:00 - Introduction 2:14 - Physical Intelligence 3:47 - Key challenges in robotic learning 6:13 - Reinforcement learning in π0 and robotic foundation models 8:36 - π0 VLM model architecture 15:33 - π0 model recipe 18:39 - Pre-training dataset 22:47 - Post-training 24:23 - Laundry folding demo 31:32 - Scaling laws on π0 model 34:57 - FAST 40:26 - Open sourcing π0 43:37 - Other robot types 46:27 - Future directions

The TWIML AI Podcast

19,942 次观看 • 1 年前

If your financial model doesn’t tell you how sensitive your outcomes are to key assumptions, you’re only seeing half the picture. That’s where Sensitivity Analysis steps in. Whether you're modeling EBITDA, company valuation, or cash flow, testing the impact of changes in assumptions—like growth rate, margin, or discount rate—is crucial for informed decision-making. 📊 3 Powerful Tools in Excel for Sensitivity Analysis: 1️⃣ Data Tables Easily test the impact of one or two input variables on a key output. Perfect for: Valuation scenarios, breakeven analysis, IRR stress tests. 2️⃣ Scenario Manager Create and store multiple input sets and compare outputs. Perfect for: Best case / base case / worst case planning. 3️⃣ Manual What-Ifs with Named Ranges + Excel Formulas Use dynamic formulas, dropdowns, and structured inputs to test anything. Perfect for: Custom, presentation-ready models with clear audit trails. 💡 Pro Tip: Combine sensitivity analysis with clear data visualization (charts, conditional formatting, slicers) to drive your story home. Finance isn't just about building the model—it's about telling the right story with the right levers. 👇 How do you use sensitivity analysis in your day-to-day modeling? HERE IS A VIDEO

If your financial model doesn’t tell you how sensitive your outcomes are to key assumptions, you’re only seeing half the picture. That’s where Sensitivity Analysis steps in. Whether you're modeling EBITDA, company valuation, or cash flow, testing the impact of changes in assumptions—like growth rate, margin, or discount rate—is crucial for informed decision-making. 📊 3 Powerful Tools in Excel for Sensitivity Analysis: 1️⃣ Data Tables Easily test the impact of one or two input variables on a key output. Perfect for: Valuation scenarios, breakeven analysis, IRR stress tests. 2️⃣ Scenario Manager Create and store multiple input sets and compare outputs. Perfect for: Best case / base case / worst case planning. 3️⃣ Manual What-Ifs with Named Ranges + Excel Formulas Use dynamic formulas, dropdowns, and structured inputs to test anything. Perfect for: Custom, presentation-ready models with clear audit trails. 💡 Pro Tip: Combine sensitivity analysis with clear data visualization (charts, conditional formatting, slicers) to drive your story home. Finance isn't just about building the model—it's about telling the right story with the right levers. 👇 How do you use sensitivity analysis in your day-to-day modeling? HERE IS A VIDEO

Bojan Radojicic

14,714 次观看 • 4 个月前

From Palantir and Two Sigma to turning mechanistic interpretability into a production-grade platform, Mark Bissell (Member of Technical Staff) and Myra Deng (Head of Product) are building Goodfire on the belief that “understanding what’s inside the model” is the path to safer, more controllable AI and Goodfire just raised a $150M Series B at a $1.25B valuation to scale that bet. We sat down with Mark and Myra to unpack what actionable interpretability looks like beyond demos: why SAE features can lose to raw-activation probes on real detection tasks, how they deployed token-level PII monitoring with Rakuten across English + Japanese without training on real customer data, and what it takes to do real-time steering on a trillion-parameter model (including the infamous Gen-Z feature). We also dig into hallucination detection, surgical post-training edits for bias/unlearning, the emerging equivalence between activation steering and in-context learning, and their push to bring interpretability into training so model development isn’t just “data in, weights out.” Goodfire mark bissell Myra Deng

From Palantir and Two Sigma to turning mechanistic interpretability into a production-grade platform, Mark Bissell (Member of Technical Staff) and Myra Deng (Head of Product) are building Goodfire on the belief that “understanding what’s inside the model” is the path to safer, more controllable AI and Goodfire just raised a $150M Series B at a $1.25B valuation to scale that bet. We sat down with Mark and Myra to unpack what actionable interpretability looks like beyond demos: why SAE features can lose to raw-activation probes on real detection tasks, how they deployed token-level PII monitoring with Rakuten across English + Japanese without training on real customer data, and what it takes to do real-time steering on a trillion-parameter model (including the infamous Gen-Z feature). We also dig into hallucination detection, surgical post-training edits for bias/unlearning, the emerging equivalence between activation steering and in-context learning, and their push to bring interpretability into training so model development isn’t just “data in, weights out.” Goodfire mark bissell Myra Deng

Latent.Space

19,389 次观看 • 4 个月前