TensorFlow Dev Summit 2020 Keynote

Key Takeaways

[Music] hi everyone welcome to the 20/20 tensorflow developer summit livestream I'm Megan cojolya VP of engineering for tensorflow thanks for tuning in to our fourth annual developers summit in our first ever virtual event with the recent developments of the coronavirus for wishing all of you good health safety and well-being well we can't meet in person we're hoping the dev summit is more accessible than ever to all of you we have a lot of great talks along with exciting announcements so let's get started when we first open source tensorflow our goal was to give everyone a platform to build AI to solve real world problems I'd like to share an example of one of those people Erwin is a radiologist in the Philippines and no stranger to bone fracture images like the ones that you see here he's a self-proclaimed AI enthusiast and wanted to learn how a I could be applied to radiology but was discouraged because he didn't have a computer science background but then he discovered tensorflow j/s which allowed him to build this machine learning application they could classify bone fracture images now he hopes to inspire his fellow radiologists to actively participate in building AI to ultimately help their patients and our wounds not alone tensorflow has been downloaded millions of times with new stories like Owens popping up every day and it's a testament to your hard work and contributions to making tensorflow what it is today so on behalf of the team I want to say a big thank you to everyone in our community taking a look back 2019 was an incredible year for tensorflow we certainly accomplished a lot together we kicked off the year with our dev summit launched several new libraries and online educational courses hosted our first google Summer of Code went to 11 different cities for the tensorflow Roadshow and hosted the first tensorflow world last fall 2019 was also a very special year for tensorflow because we launched version 2.0 it was an important milestone for the platform because we looked at tensorflow end-to-end and asked ourselves how can we make it easy to use some of the changes were simplifying the API settling on Keros and eager execution and enabling production to more devices the community really took the changes to heart and we've been amazed by what the community is built here are some great examples from winners of our 2.0 DEF post challenge like disaster watch a crisis mapping platform that aggregates data and predicts physical constraints caused by a natural disaster or deep pavlov an NLP library for dialogue systems and like always you told us what you liked about the latest version but more importantly what you wanted to see improved your feedback has been loud and clear you told us that building models is easier but that performance can be improved you also are excited about the changes but migrating your one dot X system titude o is hard we heard you and that's why we're excited to share the latest version tensorflow - not - we're building off of the momentum from - dot o last year you've told us speed and performance is important that's why we've established a new baseline so we can measure performance in a more structured way for people who have had trouble migrating - to work making the rest of the ecosystem compatible so your favorite libraries and models work with 2x finally we're committed to X core library so we won't be making any major changes but the latest version is only part of what we'd like to talk about today today we want to spend the time talking about the tensorflow ecosystem you've told us that a big reason why you love tensorflow is the ecosystem it's made up of libraries and extensions to help you accomplish your end-to-end ml goals whether it's to do cutting-edge research or apply ml in the real world there's a tool for everyone if you're a researcher the ecosystem gives you control and flexibility for experimentation for applied ml engineers or data scientist you get tools that help your models have real-world impact finally there are libraries in the ecosystem that can help create better a Iook for your users no matter where they are all of this is underscored by what all of you the community bring to the ecosystem and our common goal of building AI responsibly will touch upon all of these areas today let's start first with talking about the tensorflow ecosystem for research tensorflow is being used to push the state of the art of machine learning in many different subfields for example natural language processing is an area where we've seen tensorflow really help push the limits in model architecture the t5 model on the Left uses the latest in transfer learning to convert every language problem into a text a text format the model has over 11 billion parameters and was trained off of the colossal clean crawled corpus data set meanwhile Mena the conversational model on the right has over two point six billion parameters and is flexible enough to respond sensibly to conversational context both of these models were built using tensorflow and these are just a couple examples of what tensorflow is being used for in research there are hundreds of papers and posters that were presented at nerves last year that use tensor flow we're really impressed with the research produced at tensorflow every day at Google and outside of it and we're humbled that you trust tensor flow with your experiments so thank you but we're always looking for ways to make your experience better I want to highlight a few features in the ecosystem that will help you in your experiments first we've gotten a lot of positive feedback from researchers on tensor board dev a tool we launched last year that lets you upload and share your experiment results by URL the URL allows for quickly visualizing hyper parameter sweeps and nerves we were happy to see paper starting to cite tensor board dev URLs so that other researchers could share experiment results a second we're excited to introduce a new performance profiler tool set in tensor board that provides consistent monitoring of model performance we're hoping researchers will love the toolset because it gives you a clear view of how your model is performing including in-depth debugging guidance you'll get to hear more about tensor board dev and the new profiler from Gaul and Schumann's talks later today researchers have also told us at the changes than two decks make it easy for them to implement new ideas changes like eager execution in the core it supports numpy arrays directly just like all the packages in the PI data ecosystem you know and love the TF data pipelines we rolled out are all reusable make sure you don't miss Rohan's TF data talk today for the latest updates and tensor flow data sets are ready right out of the box many of the data sets you'll find were added by our google Summer of Code students so I want to thank all of them for contributing this is a great example of how the TF ecosystem is powered by the community finally I want to round out the tensor flow ecosystem for research by highlighting some of the add-ons and extensions that researchers love libraries like TF probability and TF agents work with the latest version and experimental libraries like Jax from Google research are composable with tensor flow like using tensor flow data pipelines to input data into Jack's but tensor flow has never just been about pushing the state of the art in deep learning a model is only as good as the impact it has in the real world this is one of tensorflow core strengths it has helped AI scale to billions of users we've seen incredible ml applications being built with tensor flow we're really humbled by all the companies big and small who trust tensor flow with their ml workloads going from an idea to having your AI create real world impact can be hard but our users rely on tensor flow to help them accomplish this that's because the tensor flow ecosystem is built to fit your needs it makes having to go from training to deployment less of a hassle because you have the libraries and resources all in one platform there's no switching costs involved I want to highlight a few new things that will help you get to production faster first you told us that you love working with Karis and tensorflow because it's easy to build and train custom models so we're committed to keeping EF Karis the default high-level API but if you're not looking to build models from scratch tensorflow hub hosts all the ready to use pre trained models in the ecosystem there are more than a thousand models available in TF hub with documentation code snippets demos and interactive collapse all ready to be used when you're ready to put your model into production you can build production ready AML pipelines in tensorflow extended to make sure your ml engineering just works from data validation to ML metadata tracking and today I'm very excited to announce that using tensorflow in production is getting even easier with an exciting launch Google Cloud AI platform pipelines we've partnered with Google cloud to make it easy to build IND in production pipelines using coop flow and tensorflow extended hosted by Google cloud cloud AI platform pipelines are available today in your Google Cloud console and if you're running tensorflow on Google Cloud tensorflow Enterprise which we announced last year at TF world gives you the long-term support and the enterprise scale that you need finally you can train and deploy your models and pipelines on custom hardware specifically designed for AI workloads cloud TP use in the latest version tensorflow is now optimized for cloud CPUs using Kerris this means the same api you started with now helps you scale to petaflop of PTP you compute all of these libraries are within the tensor flow ecosystem our 2.2 compatible and help you scale so your email application can reach your users but for AI to have that kind of impact it needs to be where your users are which means getting your models on device now we all know this requires working in some constraints like low latency working with poor network connectivity all while trying to preserve privacy you can do all of this by using tools within the tensor flow ecosystem like tensorflow light which can help make your models run as fast as possible whether it's on CPUs GPUs DSPs or other accelerators here's an example of how we've optimized performance for mobile net be one from May last year to today it's a big reduction in latency and something you get right out of the box with TF light we're also adding Android studio integration so you can deploy models easily just simply drag and drop into Android studio and automatically generate the Java classes for the TF light model with just a few clicks when network connectivity is a problem and you need these power intensive models to work while still offline you can convert them to run better on device using tensorflow light in the latest version we rebuilt the TF light converter from the ground up to provide support for more models more intuitive error messages when conversions fail and support for control flow operations the browser has become an exciting place for interactive ml and tensorflow Jas is allowing JavaScript and web developers to build some incredible applications there's some exciting new models that are now supported like Facemash and mobile Bert hugging face introduced a new NPM package for T fjs which allows you to do question answering directly in nodejs finally the new web assembly back-end is available for improved CPU performance the next few years will see an explosion of platforms and devices for machine learning and the industry needs a way to keep up MLI r is a solution to this rapidly changing landscape its compiler infrastructure for TF and other frameworks and it's backed by 95% of the world's hardware accelerator manufacturers and it's helping to move the industry forward we see how important infrastructure like ml ir is to the future of ML which is why we're investing in the future of tensorflow zone infrastructure the new tensorflow runtime is something you won't be exposed to as a developer or researcher but it will be working under the covers give you the best performance possible across a wide variety of domain-specific hardware we're planning on integrating the new runtime later this year but you'll hear more from Ming Shan later today so to recap everything you've seen so far whether you're pushing the state of the art and research applying ml to real-world problems or looking to deploy AI wherever your users are there's a tool for you in the tensor flow ecosystem now I'd like to invite mana C on stage to talk about how the ecosystem is helping empower responsible AI thank you thanks Megan hi everyone my name is mano C Joshi and I'm an engineering director on tensorflow team as Megan mentioned and you saw tensorflow ecosystem is composed of number of useful libraries and tools that are used that are useful for a diverse set of use cases whether they are coming from MLT searchers or practitioners alike however the field of ml ml and AI is raising the question whether we are building systems in the most inclusive and secure way I'm here to tell you how tensorflow ecosystem empowers its users to build systems responsibly and moreover what type of tools and resources are available to our users to accomplish those goals before I mean deep dive into the details of what tends to flow has to offer its users let's take a step back and define what we mean by responsible AI as we know that machine learning has tremendous power for solving lots of challenging real-world problems however we have to do this responsibly now to us intends to flow the way we define responsible AI is based on a five pillars strategy number one general recommended practices for AI this is all about reliability all the way from making sure that your model is not overfitting to your training data it is more generalized than that making sure you are aware of limitations of your training data when it comes to different feature distribution ski for example ensuring that the model outputs are robust when the training data gets perturbed ensuring you're not using only a single metric across all your models to determine its quality because different metrics matter to different context how your model is used for promotion demotion and filtering ranking so on and so forth right the second principle fairness fairness is a fairly evolving thing in AI for us we define it as not to create or reinforce unwanted bias fairness can be extremely subjective can be context sensitive and it is a social socio technical challenge third interpretability interpretability is all about understanding the mechanics or behind models prediction ensuring that you understand what features really matter to the final output which features what important which features were not fourth privacy for us this is all about taking into account sensitivity of your training data and features and fifth is security in the context of ML security really means that you understand vulnerabilities in your system and the threat models that are associated now for a typical user of tensorflow this is how the overall developer workflow looks like you start by defining a specific goal and objective for why you want to build the system then you go about gathering relevant data for training your model as we understand data is gold for machine learning model training and so you have to prepare the data well you have to transform it you have to cleanse it sometimes and then once your data is ready you go about training your model once the model is built its converged you then go about deploying the model in production systems that want to make use of an hour deployment phase is not a one-time task you have to continuously keep iterating in an ml workflow and improving the quality of the model now along this developer workflow there are many many different moments at which you as a modeler needs to be asking all of these questions questions like who is the audience for my machine learning model who are the stakeholders and what are the individual objectives for the stakeholders going onto the data side of it is my data really representing real-world biases or distribution skews and do I understand those limitations am I allowed to use certain features in a privacy-preserving way or they are just simply not available due to constraints then on to the training side of it do I understand implications of the data on model outputs or not a mind you doing deployments very blindly or am I being little bit mindful about deploying only reliable and inclusive models and finally when we talk about iterative workflow do I understand complex feedback loops that could be present in my modeling workflow now along all of these questions I'm happy to tell you that tensorflow eco system has few set of tools which could be helpful to answer some of them I'm not going to go through everything here but to just give you few examples starting with fairness indicators it's a very effective way by which you can evaluate your models performance across many different subgroups in a confidence interval powered way such that you can evaluate simple but effective fairness metrics for your models we have what if tool that gives you the notion of interpreting the models output based on the features and changing those features to see the changes in the models output it has very compelling textual as well as visual information associated with your data and then finally tensorflow federated it's the tensile flow to detects compatible library that helps you train your models with data that's available on device cat and Miguel have a talk later today that dives deep into fairness and privacy pillars of the responsible AI strategy be sure not to miss it we are excited to work on this important part of tensorflow ecosystem with all of you the tensorflow community and now to talk more about the community I would like to turn it over to Kemal thank you thank you - e hi everyone my name is Kemal and I'm the product director for tensorflow so you've heard a lot of from Meghan and mana C about our latest innovations now I'm going to talk about the most important part of what we're building and that's the community and I want to begin by thanking all of you your feedback your contributions what you build this is what makes all of this possible we have an incredible global community we love hearing from you and we really appreciate everyone that came out to a Roadshow of tensorflow world last year and going into 2020 I want to take some time to highlight more opportunities to get involved in the community and new resources to help you all succeed let's start with ways you can get involved locally one great way to connect is to join a tensile full user group these grassroot communities started organically and we now have 73 of them globally we launched the first two in thatin America after the roadshow in San Paulo and now we expanded our presence in Europe the Korea group is the biggest with forty six thousand members and China has user groups in 16 cities I'm sure this map can have a lot more dots so if you want to start a user group please reach out and we'll help you get started another way to get involved are the special interest groups or SIG's SIG's exists to help you build areas of tensorflow that you care the most about we will have 12 six with six graphics being our latest addition starting at the end of the month most SIG's are led by members of the open-source community such as our fantastic Google developer experts we love our gd's we now have 148 of them and I want to take a moment to recognize all of them they give Tech Talks organize workshops and dock sprints and I want to give a special shout-out to haste on pictured above who organizes doctrine in Seoul they reviewed several PRS and wrote hundreds of comments in five hours again gd's are amazing so please let us know if you're interested in becoming one okay so the tensorflow user group cigs and GDS are great ways to get involved but we all love a little competition and we all love kaggle as Kegel now supports two point acts we've had over a thousand teams enrolled in our last competition I want to give a special shout-out to our 2.0 Prize winner deep thought and speaking of competition we saw great projects in our last death Bush challenge including psychopathology assistant an intelligent assistant that tracks patients responses during face-to-face and remote sessions and everybody everyone dance faster and everybody dance now video generation library using HTTP you intensify 2.0 thank you to everyone who participated and today we have a new challenge so - II spoke earlier about how tensorflow can help empower all users to build AI systems responsibly so we want to challenge you to create something great with tensorflow 2.2 and something that has the e aí principles at heart we can't wait to see what you build so another area that we're investing in a lot is education starting with our supporting our younger community members for the first time we participated in google coding and it was a success we were very impressed by the students and we want to thank all the awesome mentors who made this possible we hope someday to see the students at our Summer of Code program I love Summer of Code it's an awesome opportunity for students to work with tensorflow engineers we saw amazing projects in fact one of the students worked on data visualization for Swift which is still being used today by our team some happy announced we're doing it again this summer and we're excited to see what new projects students will work on programs like this are to the growth of the developer community so please visit the Summer of Code website to learn more and apply we also want to help produce great educational content starting with our machine learning crash course a great resource for beginners so today we launched an updated version of the course our ngu team completely revamped the programming exercises using tariff Kerris 2.0 and made them much simpler in the process go check it out on this link and we want to provide resources at every stage of learning at the university level we want to empower educators and support them as the design develop and teach machine learning courses last year we supported Georgia Tech the University of Hong Kong Pace University and many others and this year we have a commitment to fund schools from underrepresented communities in AI historically black and Latin acts colleges and universities so if you're a faculty and you want to teach ml please reach out and we also want to help people self-study that's why we partner with deep learning about AI to give people access to great educational material today over 200,000 people have enrolled in our courses the data and deployment course is a great specialization course that covers 10 so far Gia's tensorflow Lite tensorflow dataset and more advanced scenarios this is a great option for people who are really looking to build their ml coding skills and you can audit it for free and there's more Imperial College London just released a getting started with tencel full course on Coursera this course was created in part by the tencel for funding I mentioned earlier and we're super happy to see so you're taking all these courses you're becoming vetted ml but how do you show your expertise to the world this is why I'm excited to announce the launch of the tensorflow certificate program an assessment created by the tensorflow team covering topics such as tech certification using NLP to build spam filters computer vision using CNN to do image recognition sequences and prediction bypassing this foundational certification you be able to share your expertise with the world and display your certificate badge on LinkedIn github or the tens of full certificate network and to widen access to people of diverse backgrounds and experiences were excited to offer a limited number of stipends for covering the certification costs you can find out more at tensorflow org slash certificate so a lot of things to do and I want to thank you again for making the tensorflow community so awesome as you've seen the ten salute ecosystem is having incredible impact in the world today and what it really comes down to is how AI is helping make people's lives better that's really what inspires us as a team to build all these amazing tools so I'd like to end by sharing one final story mid-south hotel is providing care in 70 countries each time we go in the new country in new wars on we discover more and more patients infected with multi drug-resistant bacteria they're arriving with the wrong antibiotics even to them antibiotic resistance is everywhere if we don't act today we'll have 10 million deaths per year by 2050 once you identify that there is a bacteria you do multiple tests to know which type of bacteria it is and then you test the sensitivity of several antibiotics in order to identify the most effective one there is an additional step that is the interpretation of those results in the majority of the countries where we work there is a lakh microbiologist and human resources to do this interpretation step n where he is ten year olds and his injury was bullet it exploded inside his leg he has very resistant type of infection we are the last hope for anyway I thought AI might be one of the solution to this problem we developed an applications who helped lab technicians to interpret the results of diagnosis tests only using their mobile phones we're using tensorflow computer vision and machine learning to detect interactions happening between the bacteria and antibiotics based on an app loaded image of the petri dish our goal is not to replace the lab technicians the app is really meant to support them and doing their diagnosis tests whilst always keeping a human in the loop we managed to train a model within a matter of days using Kara's and 15,000 anonymize pictures of diagnosis tests the API was very expressive and it was surprisingly really quick and easy to achieve we are deploying the app using tens of flow lights so I can be used offline on any range of mobile devices in all of our clinics today we have a prototype we've got a responsibility to make the diagnostic tests available easy affordable all over the world I'm excited because for me this application can be a game-changer that's going to help million and million of liván for anymore there is one antibiotic suitable for disinfection I think this up that will solve many problems everywhere in the world [Music] that's just amazing and incredibly inspiring when I see something like this it makes me very proud to be building tensorflow so go build amazing things and we'll be there to help and with that I will pass it on to page to kick off our day thank you [Music]