- Ai - Engineering Mailing Lists

[CoRIS-List] Reminder: OSU Robotics Seminar, Starship Robotics, April 15
by Ross L. Hatton 14 Apr '22

14 Apr '22

Hi All, This week, the OSU Robotics Seminar will be hosting a talk by Dylan Barrett (replacing Chris Neider) of Starship Robotics (the company behind the food delivery robots on campus). Dylan has launched Starship's delivery services at several large campuses, and manages their operations across a range of campuses with different sizes and needs. He will be presenting on both Starship's operating model and on the technology that makes it possible, and will have a robot with him at the seminar. The seminar will be an *in person* event, simulcast over Zoom, and will be immediately followed by a half-hour students-only Q&A session with the speaker. April 15, 10am LINC 268 https://oregonstate.zoom.us/j/93908195212?pwd=cEdoQVdac0JxMUpOSS9xVzZjVG5xU… =========== Ross L. Hatton Associate Professor, Robotics and Mechanical Engineering Collaborative Robotics and Intelligent Systems Institute Oregon State University coris.oregonstate.edu research.engr.oregonstate.edu/lram/ rosslhatton.com ross.hatton(a)oregonstate.edu

1 0

AI seminar: April 20, 2022
by Mangannavar, Rajesh Devaraddi 14 Apr '22

14 Apr '22

Dear all, Our next AI seminar on *"**Upscaling Nonlinear Reactions in Tissues: Closure via Machine Learning**" *by professor Brian Wood is scheduled to be on April 20th, 1-2 PM PST. It will be followed by a 30-minute Q&A session by the graduate students. Zoom Link: https://oregonstate.zoom.us/j/93591935144?pwd=YjZaSjBYS0NmNUtjQzBEdzhPeDZ5U… *Upscaling Nonlinear Reactions in Tissues: Closure via Machine Learning* Brian Wood Professor Chemical, Biological, and Environmental Engineering Oregon State University *Abstract:* In this work, we use a combination of formal upscaling and data-driven machine learning for explicitly closing a nonlinear transport and reaction process in a multiscale tissue. The classical effectiveness factor model is used to formulate the macroscale reaction kinetics. We parsimoniously train a multilayer perceptron network using training data generated by direct numerical simulations over thousands of microscale examples. Once trained, the network is applied in an algorithm for numerically solving the upscaled (coarse-grained) differential equation describing mass transport and reaction in two example tissues. The network is described as being explicit in the sense that the network is trained using macroscale concentrations and gradients of concentration as components of the feature space. Network training and solutions to the macroscale transport equations were computed for two different tissues. The two tissue types (brain and liver) exhibit markedly different geometry and spatial scale (cell size and sample size). The upscaled solutions for the average concentration are compared with numerical solutions derived from the microscale concentration fields by a posteriori averaging. There are two outcomes of this work of particular note: 1) we find that that the trained network exhibits good generalizability, and it is able to predict the effectiveness factor with high fidelity for realistically-structured tissues despite the significantly different scale and geometry of the two example tissue types; and 2) the approach results in an upscaled PDE with an effectiveness factor that is predicted (implicitly) via the trained neural network. This latter result emphasizes our purposeful connection between conventional averaging methods with the use of machine learning for closure; this contrasts with some machine learning methods for upscaling where the exact form of the macroscale equation remains unknown. *Speaker Bio:* Brian Wood specializes in transport phenomena at multiple scales. His research usually integrates theory, numerical simulation, and experimental data. Systems that include a biological component has been a particular focus, and recent work has involved the multiscale modeling of tissues with closures informed by machine learning. Wood has a significant and evolving interest in parsimoniously using deep neural networks to help “close” nonlinear problems in formal upscaling (e.g., multi-matieral mechanics, turbulence in fluid mechanics, etc.). The connections between thermodynamics and statistical mechanics is a personal interest; he is actively involved in research to understand how entropy, information, and complexity connect in physical systems. *Please watch this space for future AI Seminars :* * https://eecs.oregonstate.edu/ai-events <https://eecs.oregonstate.edu/ai-events>* Rajesh Mangannavar, Graduate Student Oregon State University ---- AI Seminar Important Reminders: -> For graduate students in the AI program, attendance is strongly encouraged.

1 0

AI seminar: April 13, 2022
by Mangannavar, Rajesh Devaraddi 12 Apr '22

12 Apr '22

Dear all, Our next AI seminar on *"**Human-AI Collaboration for Interactive Image Editing and Machine Autonomy**" *by professor Bolei Zhou is scheduled to be on April 13th (Tomorrow), 1-2 PM PST. It will be followed by a 30 minute Q&A session by the graduate students. Zoom Link: https://oregonstate.zoom.us/j/93591935144?pwd=YjZaSjBYS0NmNUtjQzBEdzhPeDZ5U… *Human-AI Collaboration for Interactive Image Editing and Machine Autonomy* Bolei Zhou Assistant Professor Computer Science University of California, Los Angeles *Abstract:* Deep learning models such as ConvNet and transformers have made huge progress in real-world applications from image generation to self-driving vehicles. Researchers and developers mainly focus on building larger and deeper models to improve the accuracy and numbers on public benchmarks. However, it is difficult to establish a reliable and trustworthy relationship between humans and AI if there are no meaningful interactions. In this talk, I will present our effort to facilitate the Human-AI collaboration for certain tasks. I will first introduce our work on improving the controllability of deep generative models such as GANs for interactive image editing. I will then talk about Human-AI Copilot (HACO), a new human-in-the-loop reinforcement learning method to efficiently teach agents to drive in uncertain environments while preserving their own curiosity and exploration. *Speaker Bio:* Bolei Zhou is an Assistant Professor in the Computer Science Department at the University of California, Los Angeles (UCLA). He earned his Ph.D. from MIT in 2018. His research interest lies at the intersection of computer vision and machine autonomy, focusing on enabling interpretable and trustworthy human-AI interaction. He has developed a number of widely used interpretation methods such as CAM and Network Dissection, as well as computer vision benchmarks Places and ADE20K. He is an associate editor for Pattern Recognition and has been area chair for CVPR, ICCV, and AAAI. He received MIT Tech Review's Innovators under 35 in Asia-Pacific Award. More about his research is at https://boleizhou.github.io/. This talk will be followed by a talk on the topic "*Upscaling Nonlinear Reactions in Tissues: Closure via Machine Learning*" by professor Brian Wood on 20th April (Wednesday) 1-2 PM, 2022 *Please watch this space for future AI Seminars :* * https://eecs.oregonstate.edu/ai-events <https://eecs.oregonstate.edu/ai-events>* Rajesh Mangannavar, Graduate Student Oregon State University ---- AI Seminar Important Reminders: -> For graduate students in the AI program, attendance is strongly encouraged.

1 0

[CoRIS-List] OSU Robotics Seminar, Starship Robotics, April 15
by Ross L. Hatton 12 Apr '22

12 Apr '22

Hi All, This week, the OSU Robotics Seminar will be hosting a talk by Chris Neider of Starship Robotics (the company behind the food delivery robots on campus). The seminar will be an *in person* event, simulcast over Zoom, and will be immediately followed by a half-hour students-only Q&A session with the speaker. April 15, 10am LINC 268 https://oregonstate.zoom.us/j/93908195212?pwd=cEdoQVdac0JxMUpOSS9xVzZjVG5xU… =========== Ross L. Hatton Associate Professor, Robotics and Mechanical Engineering Collaborative Robotics and Intelligent Systems Institute Oregon State University coris.oregonstate.edu research.engr.oregonstate.edu/lram/ rosslhatton.com ross.hatton(a)oregonstate.edu

1 0

No AI reading group meeting
by Quintin Pope 08 Apr '22

08 Apr '22

Hi everyone, There won’t be a reading group meeting today. I’ve been busy recently, but hope to resume next week. All the best, Quintin Sent from my iPhone

1 0

[CoRIS-List] No OSU Robotics seminar this week; RGSA roundtable instead
by Ross L. Hatton 07 Apr '22

07 Apr '22

Hi All, There is no OSU Robotics seminar this week; instead the RGSA is hosting a roundtable event with the faculty about the robotics program. This meeting will be in LINC 268 at 10am Friday 4/8, and its agenda has been sent to the RGSA mailing list. -Prof. Hatton =========== Ross L. Hatton Associate Professor, Robotics and Mechanical Engineering Collaborative Robotics and Intelligent Systems Institute Oregon State University coris.oregonstate.edu research.engr.oregonstate.edu/lram/ rosslhatton.com ross.hatton(a)oregonstate.edu

1 0

AI seminar: April 13, 2022
by Mangannavar, Rajesh Devaraddi 07 Apr '22

07 Apr '22

Dear all, Our next AI seminar on *"**Human-AI Collaboration for Interactive Image Editing and Machine Autonomy**" *by professor Bolei Zhou is scheduled to be on April 13th, 1-2 PM PST. It will be followed by a 30 minute Q&A session by the graduate students. Zoom Link: https://oregonstate.zoom.us/j/93591935144?pwd=YjZaSjBYS0NmNUtjQzBEdzhPeDZ5U… *Human-AI Collaboration for Interactive Image Editing and Machine Autonomy* Bolei Zhou Assistant Professor Computer Science University of California, Los Angeles *Abstract:* Deep learning models such as ConvNet and transformers have made huge progress in real-world applications from image generation to self-driving vehicles. Researchers and developers mainly focus on building larger and deeper models to improve the accuracy and numbers on public benchmarks. However, it is difficult to establish a reliable and trustworthy relationship between humans and AI if there are no meaningful interactions. In this talk, I will present our effort to facilitate the Human-AI collaboration for certain tasks. I will first introduce our work on improving the controllability of deep generative models such as GANs for interactive image editing. I will then talk about Human-AI Copilot (HACO), a new human-in-the-loop reinforcement learning method to efficiently teach agents to drive in uncertain environments while preserving their own curiosity and exploration. *Speaker Bio:* Bolei Zhou is an Assistant Professor in the Computer Science Department at the University of California, Los Angeles (UCLA). He earned his Ph.D. from MIT in 2018. His research interest lies at the intersection of computer vision and machine autonomy, focusing on enabling interpretable and trustworthy human-AI interaction. He has developed a number of widely used interpretation methods such as CAM and Network Dissection, as well as computer vision benchmarks Places and ADE20K. He is an associate editor for Pattern Recognition and has been area chair for CVPR, ICCV, and AAAI. He received MIT Tech Review's Innovators under 35 in Asia-Pacific Award. More about his research is at https://boleizhou.github.io/. This talk will be followed by a talk on the topic "*Upscaling Nonlinear Reactions in Tissues: Closure via Machine Learning*" by professor Brian Wood on 20th April (Wednesday) 1-2 PM, 2022 *Please watch this space for future AI Seminars :* * https://eecs.oregonstate.edu/ai-events <https://eecs.oregonstate.edu/ai-events>* Rajesh Mangannavar, Graduate Student Oregon State University ---- AI Seminar Important Reminders: -> For graduate students in the AI program, attendance is strongly encouraged.

1 0

AI seminar: April 6, 2022
by Mangannavar, Rajesh Devaraddi 05 Apr '22

05 Apr '22

Dear all, Our next AI seminar on *"Charismatic Robots in Everyday Human Spaces" *by professor Heather Knight is scheduled to be on April 6th (Tomorrow), 1-2 PM PST. It will be followed by a 30 minute Q&A session by the graduate students. Zoom Link: https://oregonstate.zoom.us/j/93591935144?pwd=YjZaSjBYS0NmNUtjQzBEdzhPeDZ5U… *Charismatic Robots in Everyday Human Spaces* Heather Knight Assistant Professor Computer Science Oregon State University *Abstract:* The pandemic has brought increasing numbers of robots and automation into everyday human spaces, making our prior work in service robots, expressive communication, and autonomous and human-in-the-loop robot behavior systems ever more relevant. Using examples from 20 years in the field, this talk illustrates how technology designers can repeatedly leverage human-inspired communication systems into robot perception and enactment systems that can add social and functional value to robots operating with, for, and around humans. It will also present work from the CHARISMA Robotics lab at Oregon State University, an acronym for Collaborative Humans and Robotics: Interaction, Sociability, Machine learning and Art. From robot furniture to robot comedy, from nonverbal expressions to the cultural situation of robots operating around people in factories, workplaces, and sidewalks, CHARISMA contributes to the fields of human-robot interaction and social robotics, regularly deploying robots in naturalistic human settings and entertainment contexts. *Speaker Bio:* Heather Knight runs the CHARISMA Robotics research group at Oregon State University, which applies methods from entertainment to the development of more effective and charismatic robots. Their research interests include minimal social robots, multi-robot/multi-human social interaction, and entertainment robots. Her previous work includes a postdoc at Stanford University that explores minimal robots and autonomous car interfaces, a PhD in Robotics at Carnegie Mellon University exploring Expressive Motion for Low Degree of Freedom Robots, and M.S. and B.S. degrees in Electrical Engineering & Computer Science at Massachusetts Institute of Technology, where she developed a sensate skin for a robot teddy bear at the MIT Media Lab. Additional past work includes robotics and instrumentation at NASA's Jet Propulsion Laboratory, sensor design at Aldebaran Robotics, and nine years of producing an annual Robot Film Festival exploring positive storytelling about robots. Additional honors include a robot flower garden installation at the Smithsonian/Cooper-Hewitt Design Museum, robot comedy on TED.com, and a British Video Music Award for OK GO's "This Too Shall Pass" music video featuring a two-floor Rube Goldberg Machine. This talk will be followed by a talk on the topic "*Human-AI Collaboration for Interactive Image Editing and Machine Autonomy*" by professor Bolei Zhou on 13th April (Wednesday) 1-2 PM, 2022. *Please watch this space for future AI Seminars :* * https://eecs.oregonstate.edu/ai-events <https://eecs.oregonstate.edu/ai-events>* Rajesh Mangannavar, Graduate Student Oregon State University ---- AI Seminar Important Reminders: -> For graduate students in the AI program, attendance is strongly encouraged.

1 0

[CoRIS-List] Reminder: Today: OSU Robotics Seminar: Cristina Wilson, Human-robot teaming
by Ross L. Hatton 01 Apr '22

01 Apr '22

Hi All, As a reminder, our first seminar of the spring term will be today from Cristina Wilson of the University of Pennsylvania. ----- Title: A cognitive science approach to addressing bottlenecks in human-robot teaming during spatiotemporal search Speaker: Dr. Cristina Wilson Date/Time: Friday, April 1, 10-11 am https://oregonstate.zoom.us/j/93908195212?pwd=cEdoQVdac0JxMUpOSS9xVzZjVG5xU… Abstract: Robots are taking on increasingly important roles in human spatiotemporal search problems, e.g., planetary exploration, search and rescue, wildfire management. However, state-of-the-art robots are still struggling to support human experts’ reasoning and decision making in the type of complex and uncertain environments that characterize search problems. In this talk, I identify the common bottlenecks in human-driven search and share my work on how collaborative robots can ameliorate bottlenecks through improved cognitive interfacing with human teammates. For example, a bottleneck in the integration of new information from the environment with existing abstract search objectives can lead to inflexible and maladaptive decisions about where to explore next. Computational models that translate abstract human objectives into quantitative actions (rendered as scripts) allow mobile robot assistants to aid in real-time information search. In this manner, understandings of human cognition and behavior expectations are used to inform and complement machine intelligence, endowing robots with representations of human reasoning and decision processes, so robots can, in turn, produce explainable search behaviors to the satisfaction of human teammates. About the speaker: Dr. Wilson is currently a postdoctoral fellow studying collaborative human-robot teaming in the General Robotics, Automation, Sensing and Perception laboratory at the University of Pennsylvania. She received her PhD in cognitive psychology in 2018 from Washington State University, developing expertise in the study of human cognition and behavior. Dr. Wilson’s research emphasizes the human mind in human-robot teaming, designing robotic systems that can understand human objectives and support adaptive decision making. She works closely with domain expert end-users to understand the cognitive challenges to effective robot teaming, thereby enhancing robot systems application in their domains. =========== Ross L. Hatton Associate Professor, Robotics and Mechanical Engineering Collaborative Robotics and Intelligent Systems Institute Oregon State University coris.oregonstate.edu research.engr.oregonstate.edu/lram/ rosslhatton.com ross.hatton(a)oregonstate.edu

1 0

Alignment Reading Group Meeting
by Quintin Pope 31 Mar '22

31 Mar '22

Hello everyone, Based on the survey, it looks like Friday at 2 PM PST is the time period with the least conflicts. Again, thanks to Alex for suggesting our next discussion topic: https://www.gwern.net/Clippy > It might help to imagine a hard takeoff scenario using solely known sorts of NN & scaling effects… Below is a story which may help stretch your imagination and defamiliarize the 2022 state of machine learning. It’s a fictional timeline of AI causing an existential catastrophe after escaping from an ML experiment. I look forward to meeting you all! Join Zoom Meeting https://oregonstate.zoom.us/j/95843260079?pwd=TzZTN0xPaFZrazRGTElud0J1cnJLU… Password: 961594 Phone Dial-In Information +1 971 247 1195 US (Portland) +1 253 215 8782 US (Tacoma) +1 301 715 8592 US (Washington DC) Meeting ID: 958 4326 0079 All the best, Quintin Sent from my iPhone

1 0