• Prototyping Artificial Intelligence Models

    Online event

    Big Data and Artificial Intelligence (AI) have captivated the imagination of many companies. However, to go from raw data to benefiting the end-user can be a long marathon filled with hurdles. In this presentation, we will discuss a way of sharing AI models through rapid prototyping that benefits both end-users and data scientists.

    Trunojoyo (Atun) Anggara, Ph.D. is a contractor with a financial services company. Being a lifelong learner, he belongs to the Twin Cities Deep-Learning Study Group and Analyze This Group in the twin cities area. His past experience includes collaboration with a startup company on Natural Language Processing, academic advising, and computational catalysis for his Ph.D. in Chemical Engineering at Notre Dame.

    2
  • AI and the IP Landscape

    Online event

    Ryan Phelan, J.D., MBA, a registered patent attorney, will be speaking on AI and the IP Landscape - a topic he's also written an in-depth article on. You can read the full article and get a preview of the discussion here: https://www.marshallip.com/insights/artificial-intelligence-the-intellectual-property-landscape/

    Ryan also recently launched a Patent and Intellectual Property (IP) law blog focusing on Next-Generation and New-Age Technologies: https://www.patentnext.com/

    On Our Presenter: Ryan works with clients in all areas of intellectual property (IP), with a focus on patents. Clients enjoy Ryan’s business-focused approach to IP. With a MBA from Northwestern’s Kellogg School of Management, Ryan works with clients to achieve their business objectives, including developing and protecting their innovations and businesses with IP.

    Ryan routinely helps clients with:

    Preparing and prosecuting high-quality patent applications, and developing strategic patent portfolios for innovative products and services, including in the U.S. and foreign jurisdictions (e.g., Europe, China, and Japan).

    Preparing legal opinions as to the patentability, non-infringement, validity, and/or freedom-to-operate of innovative products or services.

    Litigating IP related issues to protect client market share from competitors and defend clients from IP lawsuits against competitors or non-practicing patent entities.
    As a former technology consultant with Accenture and with a background in computer science and engineering, Ryan has extensive experience in computer system, hardware, and software design, engineering, development and related technologies. He represents numerous startup and Fortune 500 clients with patent matters in technical areas and industries including artificial intelligence and machine learning, medical devices, biometrics data and services, virtual reality, imaging, internet and e-commerce, computer networking, data storage and management, encryption and security, mobile telecommunications, consumer electronics, insurance and finance applications, mechanical devices, among others.

    Ryan has been published in several well-known IP publications including the World Intellectual Property Review, Bloomberg Law, and IP Litigator, and has spoken as a panelist at various IP conferences, including for the International Intellectual Property Law Association (IIPLA), and for the University of Illinois Chicago Annual IP Conference. Ryan's recent article titled Artificial Intelligence & the Intellectual Property Landscape details how artificial intelligence is reshaping the business and intellectual property landscape.

    Ryan is also an adjunct professor at Northwestern University’s Pritzker School of Law where he teaches coursework on Patenting Software Inventions, with a focus on patent subject matter eligibility dealing with procuring software and computer related patents in view of 35 U.S.C. § 101 and the U.S. Supreme Court’s decision in Alice Corp. v. CLS Bank International, 573 U.S. 208 (2014).

    Ryan has been rated by Super Lawyers® and Leading LawyersSM as a Rising Star and peer selected Emerging Lawyer, respectively, for years 2019 through 2020. Since 2021, Ryan has been named a Leading Lawyer.

  • Facial Animation Generation with Applications in Educational Psychology

    Abstract:
    This talk will present preliminary work in facial animation generation with applications in educational psychology. In the first part, we describe two psychology studies as well as the computer vision techniques and platforms being used. Both studies investigate using conversational agents (CAs) as a way of delivering medical messages to patients. By incorporating CAs in the system, both semantic and emotional information can be delivered, which helps the patients, especially those with low heath and numerical literacy, to get a better understanding of their test results and medical instructions. Human studies were conducted to test the effectiveness of CA. The second part of this talk will discuss the details of a proposed neural network based facial animation synthesis method. By unifying both appearance-based and warping-based methods in an end-to-end training process, the proposed system was able to generate vivid facial animation with highly preserved details. In addition, we integrated this network with another audio speech processing system. We show both qualitatively and quantitatively that the proposed system achieved a higher performance than baseline methods. Finally, another two studies regarding representation learning using adversarial autoencoders, as well as infant gaze direction classification will be briefly reviewed.

    Bio:
    Kevin Gu is a new employee who just joined 3M CRSL last September. Kevin has a background in Electrical and Computer Engineering, with a focus on image/signal processing during Bachler and Master degree, and computer vision and deep learning in PhD degree. Before joining 3M, Kevin did a summer internship at St. Paul, where he worked with people in CRSL on QR code detection and code migration from C++ to Java on Android device.

  • Considerations for Machine Learning in the Wild

    Online event

    Moving any technology from theory to practice involves surprises, and machine learning (ML) is no exception. In this presentation, we will share lessons from applying ML in consumer, health care, manufacturing, and industrial applications. We will set the stage with a quick review of the ML pipeline. The remainder of the talk will focus sequentially on each step in the pipeline discussing practical considerations, common mistakes, and relevant tips.

    Presenters: Saber Taghvaeeyan & Hamid Mokhtarzadeh

    Saber has expertise in machine learning, time-series analysis, and sensor fusion. He enjoys developing end-to-end solutions involving data acquisition, analysis, and visualization. He has led projects in different industries including medical devices, wearable devices, intelligent consumables, food safety, and manufacturing.

    Hamid is passionate about navigation systems, estimation, and sensor fusion. He has academic, industry, and teaching experiences all in the area of positioning and navigation, sensor fusion, and software for scientific and engineering applications.

    2
  • Executing on MLOps To Build AI Products

    Online event

    Agenda

    6:30 – 6:40 Intro
    6:40 – 7:40 Presentation
    7:40 – 7:55 Q&A
    7:55 – 8:00 Adjourn

    Abstract

    Despite business investments, one of the biggest challenges to unlocking the potential of data science is moving machine learning projects past the experimental stage. MLOps is an emerging discipline to address this challenge. In this talk, we will discuss executing on MLOps to accelerate the migration of machine learning models to production.

    Presenter: Greg Hayes

    A technical leader in ML Engineering and Advanced Analytics with a strong interest in Python open source platforms. He has more than 20 years of experience leading global multi-disciplinary technology teams, and collaborating with global stakeholders/customers to identify and align on new opportunities. He has spent the last 7 years leveraging open source data science platforms to build value-added solutions, most recently on the Microsoft Azure technology stack.

  • Visualizing High-Dimensional Product Spaces: An End-to-End Approach

    Agenda

    6:30 – 6:40 Intro
    6:40 – 7:40 Presentation on Visualizing High-Dimensional Product Spaces
    7:40 – 7:55 Q&A
    7:55 – 8:00 Adjourn

    Abstract

    How do you search for products? In this meeting, Yaniv will report his latest efforts to help readers find their next book. Using deep-learning, he's created a map of all the books that encodes both genre and popularity. The project involves scraping data with Scrapy, preprocessing with SQL, modeling with Tensorflow, visualization in Javascript, and deployment with Flask on AWS. This project is a set of open-source tools that together cover all you need to go from a data project idea to a sharable proof-of-concept. We'll get a short intro to each tool and see how it all connects into one end-to-end workflow.

    Bio

    Yaniv Ben-Ami is an Assistant Professor of Economics at Carleton College where he teaches Finance, Macro, and Econometrics. He's the founder of the Twin Cities Deep-Learning Study Group, a weekly meetup for deep-learning enthusiasts. His research focuses on the visualization of high-dimensional data through the use of non-linear dimension reduction techniques. Some of his previous projects include creating visualizations for the American Time Use Survey (ATUS) and daily Covid-19 cases. Yale Ph.D. in Economics, Tel-Aviv University M.A. in Economics and B.Sc. in Computer Science and Biology; self-identifies as a programmer.

    1
  • Building AI models for sound recognition with less annotation effort

    Agenda

    6:30 – 6:40 Intro
    6:40 – 7:40 Building AI models for sound recognition with less
    annotation effort
    7:40 – 7:55 Q&A
    7:55 – 8:00 Adjourn

    Abstract
    Sound is one of the most important mediums to understand the environment around us. Identifying a sound event (such as a police siren, a dog bark, or a creaking door) leads to a better understanding of the context where the sound events occurred. An AI system that automatically recognizes sound events can understand our environment through sound as humans do. However, building a sound recognition model typically requires a large amount of carefully labeled audio data. A human annotator needs to listen to a lengthy audio recording and add text labels with their temporal information (onset and offset of an event) within a recording.

    In this talk, Bongjun will introduce his past research works presenting ways to reduce such human effort on sound event annotation. First, He will talk about deep learning models for sound event detection and classification that can be trained on less accurately annotated data which takes less time to collect. Secondly, He will also introduce a human-in-the-loop system for sound event annotation which helps human annotators collect sound events of interest quickly.

    Bio:
    Bongjun is a data scientist and member of AI Lab at 3M in Minnesota. He completed his PhD in Computer Science at Northwestern University as a member of the Interactive Audio Lab. His research interests include machine learning, audio signal processing (e.g., sound event recognition), intelligent interactive system, multimedia information retrieval, and human-in-the-loop interface. He also enjoys working on a musical interface and interactive media art.

    Website: https://www.bongjunkim.com/

    2
  • Using AI & Quantum Computing to Optimize Natural Resources

    Agenda

    6:30 – 6:40 Intro
    6:40 – 7:30 Using AI & Quantum Computing Presentation by Danika Hannon
    7:30 – 7:45 Q&A
    7:45 – 8:00 Wrap up

    Using AI & Quantum Computing with Boltz.ai

    During this talk, Danika Hannon will share the Boltz.ai story and talk about how she and her team are using AI and quantum computing to help farmers optimize their resource use so they can lower their operating costs.

    The Presenter

    Danika Hannon - https://www.linkedin.com/in/danikahannon/

    In her role as Head of Operations with Boltz.ai, co-organizer of the MN Quantum Computing Meetup and Senior Advisor on the Women in Quantum Advisory Board, Danika Hannon is an active part of the quantum computing community.

    Boltz.ai is a data-driven decision-support platform that optimizes resource usage to help farmers reduce operating costs for optimal growth.

    https://www.boltz.ai/

  • 2nd Learning Workshop

    Online event

    Agenda

    9:30 - 9:35 Welcome

    Learning Track Zoom Room - 2nd Workshop (Part I)
    1) Questions from 1st workshop
    2) Train the model to detect whether a person is
    wearing a face mask

    Short Break

    Learning Track Breakout - 2nd Workshop (Part II)
    1) Train the model to detect whether a person is
    wearing a face mask
    2) Discuss ways to advance the project further than
    what is covered in the workshop

    12:00 Adjourn

  • IoT Hack Day 2020!

    Online event

    We regret to inform everyone that our previously scheduled IoT Hackday Project Track has been canceled. We will however continue as planned with the Learning Track. Please join the Zoom meeting on Oct 24th at 10:30 to participate in a 2-series workshop hands-on workshop.

    -- Learning Track --
    This track is for those who are new to hacking, or don’t have a project or are not in a team.

    Project Description:
    Because of COVID-19, it is not hard to imagine that some people will prefer wearing masks for the foreseeable future to protect themselves and others. Additionally, local governments are mandating mask wearing to reduce the spread of COVID-19.

    In a 2-series workshop, we will explore TensorFlow and how it can be used to detect whether one is wearing a face mask.
    --1st Workshop--
    We will set up TensorFlow and use a pre-trained model for image classification.
    --2nd Workshop--
    We will train our model to detect whether a person is wearing a face mask.
    The instructor is Chris Black. Learn more about Chris at https://www.blackcj.com/blog/about/

    For this exercise, you need the following tools:
    * A laptop with a webcam
    * A face mask that covers your nose and mouth

    --Today - Sat, Oct 24th--
    --Project Track--

    *Submit your project at https://iothackday2020.devpost.com/. See examples of 2019 IoT Hack Day project submissions at https://iothackday2019.devpost.com/project-gallery

    *Refine your project and form/recruit your team. Or request to join a team.

    --Learning Track--
    *Sign-up for the Learning Track here https://iothackday2020.devpost.com/
    *Take several masked & maskless photos for the pre-trained model. We will share a link to submit photos that the instructor will use to pre-train the model.

    Check back for more details about prizes and team formation.

    1