This interactive session targets "newbies" to machine learning, especially students, emphasizing the breadth and depth of ML and the interdisciplinary applications possible. Participants will have the opportunity to walk through the process of decision-making to solving (by ML).
AI and SocietyComputer Vision
Engaging with digital imagery has become fundamental to participating in contemporary society. However, most digital images remain inaccessible to the 39 million people worldwide who are blind. The speakers share their perspectives as people who are both technology experts and are themselves blind.
This session aims to connect and advocate for physicists in data science. We will be discussing cutting-edge data science applications in physics research, data science education approaches in the physics curriculum, as well as how physicists with AI/ML experience can transfer their skill sets to careers in industry.
AI and Society
This panel will focus on the latest developments in the fight to slow climate change with an emphasis on the application of AI and ML, the implications of energy policy on deep learning and how GANs can help visualize climate change and it's impact, as well as high impact opportunities to get involved.
In 2018, the Global Slavery Index found that there were 40.3 M people in modern slavery. The United Nations demands companies to take immediate measures and state clear policies. The Future Society is an independent nonprofit think-and-do tank curating an up-to-date repository of >16K of those statements.
This session of TWiMLfest explores how AI is used in industrial applications. Arnab Biswas and Kai Lichtenberg discuss their experience applying AI in industrial use cases and how it differs from areas more akin to IT, followed by two use cases: Predictive Maintenance and Predictive Quality.
AI and SocietyThe Future of AI
This workshop targets teachers, parents or anyone interested in teaching AI for kids. We'll present some methods and continue with a discussion about methods, ressources and challenges when teaching AI for kids. Ever wondered where to start when teaching AI to kids with no prior coding experience? Have you already taught AI principles for kids in or out off school and want to share your experiences? We want to present two different methods/resources - more information to follow - and host a discussion to share ressources, learn from each others experiences and discuss common problems. Everyone is welcome, from the experienced AI teacher to any parent planning to teach their own children.
A session exploring the possibilities of CheckList, the task-agnostic methodology for testing NLP models introduced in the paper, Beyond Accuracy: Behavioral Testing of NLP Models with CheckList. This session is inspired by the recent TWIML interview with Sameer Singh, co-author of the paper.
Fairness and Bias in AI
Join Shalini Kantayya, Meredith Broussard, and Deb Raji for a screening of the film Coded Bias. Panelists will discuss the societal implications of the biases embedded within AI algorithms, examples of AI systems with disparate impact across industries and communities, and what can be done about it.
Bring your AI to a Codenames competition! Every contestant will submit an AI-powered bot, which will need to be able to give 1-word clues as well as interpret their teammate's. Will the winner take a word embeddings approach? A reinforcement learning approach? Something off the wall? Join us to find out!
Deep learning has become the industry standard in Computer Vision and NLP, however, in the time series domain many companies still prefer simple models. The session will go over LSTMs and other RNNs and include practical examples of how these models perform in different industries such as retail and healthcare.
Deep Learning technologies have made a tremendous impact on Medical Imaging. In this session, participants will hear about some of the latest advances in medical image segmentation and classification, as well as object detection in medical images.
As a technologist, I am gung-ho about the widespread application of technology across any domain, including healthcare. Here, I share how my conversations with doctors helped me be more nuanced about where and how AI should be applied in Healthcare.
DICOM files store image data along with a lot of metadata. Join us for a technical presentation on processing medical images stored in DICOM format before passing the data to DL algorithms, as well as the importance and usage of that metadata.
In this Keynote session, Sam is joined by friend of the show Jeremy Howard, creator of the popular fast.ai coursework. Sam and Jeremy explore topics such as the current landscape and future of AI education, as well as the new fast.ai course: Practical Deep Learning for Coders.
In this session, Sam is joined by Milind Tambe, Director of AI for Social Good at Google Research India, and Director of the Center for Research in Computation and Society (CRCS) at Harvard University. After the replay of this conversation, join Sam and Milind for an AMA!
In this Keynote Interview, we’ll explore Sal Khan’s perspective on building a large educational platform like Khan Academy, the role of community in his efforts, and how online and community education continues to evolve during the COVID19 pandemic.
In this Keynote Interview, we’re joined by Shakir Mohamed, Senior Research Scientist at DeepMind and Leader of Deep Learning at Indaba. We'll explore his journey building up the Indaba community, as well as his current research on decolonization In Artificial Intelligence.
Join this live keynote interview for a discussion with Suzana Ilić, computational linguist and founder of MLT (Machine Learning Tokyo). Sam and Suzana will explore her work applying NLP techniques to accelerating biomedical research at Causaly, as well as her experiences building the popular MLT community.
This session will provide members of the community with an overview of how to start and lead a TWIML study group. Topics include proposing a study group topic, defining your study group's scope and timeline, and best practices for encouraging participation and engagement, followed by a Q&A session.
Why is machine learning hard in the agricultural domain, how does addressing these challenges broaden our understanding of machine learning generally, and what challenges exist around adoption within the global agricultural community? This session will feature speakers from both academia and industry.
Natural Language Processing
NLP is going through a renaissance right now. How does a team shift through the noise of what is currently possible? And how does a team deliver a solution in the current market to their intended customers? Join us to find out!
Office Hours invites experts and practitioners in various topic areas for AMA (ask-me-anything) style sessions to answer technical questions from the community and/or help participants advance their specific projects and interests. This week’s topic will be centered on Computer Vision!
Natural Language Processing
Office Hours invites experts and practitioners in various topic areas for AMA (ask-me-anything) style sessions to answer technical questions from the community and/or help participants advance their specific projects and interests. This week's topic will be centered on NLP!
Office Hours invites experts and practitioners in various topic areas for AMA (ask-me-anything) style sessions to answer technical questions from the community and/or help participants advance their specific projects and interests. This week’s topic will be centered on Reinforcement Learning!
AI and the Arts
Machine learning no longer just classifies imagery, it can create new images as well. Join four students from Derrick Schultz and Lia Coleman’s Artificial Images classes as they show the work they produced as artists working with machine learning, followed by a short Q&A.
In these sessions, we invite any and all from expert to beginner to share their ML experience, knowledge, and projects. The goal is to provoke interesting project-related discussions to help the presenting and audience learn and progress their knowledge and skills. With thanks to our confirmed presenters: Madhusoodhana Chari, Nick Doiron and Nicholas Teague.
In these sessions, we invite any and all from expert to beginner to share their ML experience, knowledge, and projects. The goal is to provoke interesting project-related discussions to help the presenting and audience learn and progress their knowledge and skills. Any and all ML topic areas are welcome.
This panel will bring together AI/ML practitioners and instructors from various specializations to discuss their journey in creating artificial intelligence/machine learning courses for the community. We'll discuss how to identify the knowledge or skill gap to be addressed in your course, the steps to refine the scope and timeline of that course, useful tools and resources, best practices, and more.
AI and the Arts
StyleGAN has become one of the most popular image generation tools in the machine learning community. In this 1.5 hour workshop, join artist and educator Derrick Schultz for a brief overview of the StyleGAN2 network.
Startups and Entrepreneurship
Let us face the facts, startups don't have the money like tech giants. So, having a dedicated ML engineering team is completely out of the box. This is a story of difficulties I faced during the first few years of my career and how I overcome those. This is not much of a technical talk. If you want to hear about deep learning architecture or how I came to some aha moment because I am able to achieve higher accuracy than existing state of the art models, or how ml can help solve estimating climate change, then this talk is not for you. This is more casual tongue in cheek story of problems I encountered for past few years, pivot from struggling to build generalised ai to business impact oriented problems.