Building multimodal, integrative AI systems with Platform for Situated Intelligence

Subscribers:
344,000
Published on ● Video Link: https://www.youtube.com/watch?v=A6nr8MWwM3o



Duration: 1:13:42
1,805 views
41


In the last decade, we’ve seen fast-paced progress in many individual AI areas, such as computer vision, speech, and machine translation. However, anyone who’s tried bringing multiple AI technologies together in end-to-end systems designed for real-time, real-world interactions knows that constructing such systems remains a demanding task. Apart from research challenges that may arise in the context of a given application, the construction of multimodal, integrative AI systems is often daunting from an engineering perspective.

In this webinar, Dan Bohus, a Senior Principal Researcher in the Perception and Interaction Group at Microsoft Research, will introduce Platform for Situated Intelligence, an open-source framework that aims to address these challenges and accelerate and simplify the development, study, debugging, and maintenance of multimodal, integrative AI systems. The framework provides infrastructure for working with temporal streams of data; an efficient model for parallel, coordinated computation; rich tools for multimodal data visualization, annotation, and processing; and an open ecosystem of components that encapsulate various AI technologies. Bohus will break down the capabilities of Platform for Situated Intelligence and demonstrate how to write a very simple application using the framework, as well as how to use the available visualization tools.

Together, you’ll explore:

■ Challenges with building multimodal, integrative AI systems
■ A model for parallel, coordinated computation over temporal streams of data
■ Tools for data visualization and debugging
■ The available open ecosystem of components

𝗥𝗲𝘀𝗼𝘂𝗿𝗰𝗲 𝗹𝗶𝘀𝘁:

■ PSI Project Page: https://www.microsoft.com/en-us/research/project/platform-situated-intelligence
■ PSI on GitHub: https://github.com/microsoft/psi
■ PSI Blog on open-source framework for multimodal, integrative AI (2020): https://www.microsoft.com/en-us/research/blog/platform-for-situated-intelligence-an-open-source-framework-for-multimodal-integrative-ai
■ PSI Blog on tools and framework (2018): https://www.microsoft.com/en-us/research/blog/platform-situated-intelligence-tools-framework-multimodal-interaction-research

*This on-demand webinar features a previously recorded Q&A session and open captioning.

This webinar originally aired on September 10, 2020

Explore more Microsoft Research webinars: https://aka.ms/msrwebinars




Other Videos By Microsoft Research


2021-04-21Research for Industries (RFI) Lecture Series: Andreas Haeberlen
2021-04-13Discovering hidden connections in art with deep, interpretable visual analogies
2021-04-13ZeRO & Fastest BERT: Increasing the scale and speed of deep learning training in DeepSpeed
2021-04-13Interactive sound simulation: Rendering immersive soundscapes in games and virtual reality
2021-04-13A prototype implementation of 4G packet gateway Microsoft Catapult FPGA platform
2021-04-12Self-Tuning Networks: Amortizing the Hypergradient Computation for Hyperparameter Optimization
2021-04-06Ultra-dense data storage and extreme parallelism with electronic-molecular systems
2021-04-06Harmonizing the declarative and imperative in database systems
2021-04-06Domain-specific language model pretraining for biomedical natural language processing
2021-03-30Platform Biography: A framework for analyzing the structures and dynamics of social media
2021-03-30Building multimodal, integrative AI systems with Platform for Situated Intelligence
2021-03-29From player to creator: Designing video games on gaming handhelds with Microsoft TileCode webinar
2021-03-29Camera-based non-contact health sensing
2021-03-29Foundations of causal inference and its impacts on machine learning webinar
2021-03-29Avatars: Finding a sense of self and others in the virtual world
2021-03-25In pursuit of responsible AI: Bringing principles to practice
2021-03-25Fairness-related harms in AI systems: Examples, assessment, and mitigation
2021-03-25Enhancing mobile work and productivity with virtual reality
2021-03-23Mixed reality and robotics: Unlocking more intuitive human-machine collaboration
2021-03-23Project InnerEye: Augmenting cancer radiotherapy workflows with deep learning and open source
2021-03-23AI advances in image captioning: Describing images as well as people do