- Ai - Engineering Mailing Lists

AI Alignment Reading Group
by Pope, Quintin 11 Mar '22

11 Mar '22

Hello everyone, We'll again be meeting at 1 PM PST this Friday. We'll discuss In-context Learning and Induction Heads <https://transformer-circuits.pub/2022/in-context-learning-and-induction-hea…>. This work attempts to understand how GPT transformers are able to adapt to the current linguistic context so effectively. The authors propose a specific mechanism for in-context learning (induction heads) and argue using different approaches that induction heads account for most in-context learning. In particular, they find no evidence of mesa optimization contributing to in-context learning. If you're at all interested in the internal organization or behaviors of transformers, please feel free to attend! The paper is from the group behind A Mathematical Framework for Transformer Circuits <https://transformer-circuits.pub/2021/framework/index.html>, which we discussed previously. However, this work is more empirical and is very well explained, so I found it more approachable. 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

2 1

[CoRIS-List] OSU Robotics Seminar 3/11 (Tomorrow): Dan Oblinger
by Ross L. Hatton 11 Mar '22

11 Mar '22

Hi All, As a reminder, the final OSU Robotics seminar for the winter term will be a "fireside chat" with Dan Oblinger. He'll discuss insights from his career, which has spanned university research, work as a DARPA program manager, and founding multiple companies. The seminar will be from 10-11am, followed by an 11-11:30 student-only Q&A session. ** This seminar will be held in person in LINC 302. ** --- A simulcast of the seminar will be available via Zoom https://oregonstate.zoom.us/j/92183247338?pwd=MENOWHgvSFVGYzNjZExDT2hWRUNxd… --- -Ross =========== 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 =========== Robotics Innovation In The Lab, Gov, Corp, and Startup We use the narrative stories of Bizzy, Martian, Aeolus Robotics, touch upon Dusty Robotics, Amazon's Astro and DARPA funded work, in order to talk about robotic innovation from several vantage points. BEST PRACTICES AND ANTI-PATTERNS -- We use these examples to discuss: - Founding companies, gaining product market fit - Getting funded (both governmental and investor funding) - Growing and running larger (40 PhD) R&D robotics teams. EXAMPLES OF APPLIED RESEARCH - Low cost robotics -- approaches for reducing 2 orders of magnitude in cost - Wiping Research -- Bizzy was a bathroom cleaning robot. This involved novel on robotic path planning for surface wiping. - Unconstrained Grasping -- Aeolus was a 'helper' bot, and is required to be able to manipulate found objects in an uncontrolled setting. ROBOTICS IN THIS NEW REMOTE WORLD — Not just because of covid, technical work is moving towards a planetary teams rather than co-located teams. As you can imagine this has dramatic consequences for a robotics team. I setup a 24/7 follow the sun hardware innovation lab. It kinda worked ;-). Lessons learned and thoughts for the future of robotics. DAN OBLINGER The first third of Dan's career was focused on academic research in ML and Programming By Demonstration while at IBM research and teaching in EE at Columbia University. He then served as a DARPA Program Manager funding over $200M in R&D focused on Human like learning from instruction and Language Understanding, including the work that led to the IBM Watson system beating the Jeopardy-TV-show world champion. Recently Dan has founded and exited several companies in AI, Robotics and Clean Energy.

2 1

Machine Learning II
by Fern, Xiaoli 10 Mar '22

10 Mar '22

Hi all, Want to bring to your awareness that I will be teaching a special topic class titled Machine learning II (AI 539). The vision is that this will develop into the second course of a two-course Machine Learning series . I am attaching the syllabus for this term's offering for your consideration. Please feel free to send me an email if you have any question regarding the class. -Xiaoli Fern

3 4

AI seminar: March 9, 2022
by Mangannavar, Rajesh Devaraddi 08 Mar '22

08 Mar '22

Dear all, Our next AI seminar on *"Challenges and Opportunities in Out-of-distribution Detection" *by Sharon Yixuan Li is scheduled to be on March 9th (Tomorrow), 1-2 PM PST. *The 30-minute graduate Q&A session will be held at 11:30 AM on 10 March (Thursday).* Please note that this session is on the day after the talk. Zoom Link: https://oregonstate.zoom.us/j/93591935144?pwd=YjZaSjBYS0NmNUtjQzBEdzhPeDZ5U… *Challenges and Opportunities in Out-of-distribution Detection* Sharon Yixuan Li Assistant Professor Department of Computer Sciences University of Wisconsin-Madison *Abstract:* The real world is open and full of unknowns, presenting significant challenges for machine learning (ML) systems that must reliably handle diverse, and sometimes anomalous inputs. Out-of-distribution (OOD) uncertainty arises when a machine learning model sees a test-time input that differs from its training data, and thus should not be predicted by the model. As ML is used for more safety-critical domains, the ability to handle out-of-distribution data are central in building open-world learning systems. In this talk, I will talk about challenges, methods, and opportunities on uncovering the unknowns of deep neural networks for reliable predictions in an open world. *Speaker Bio:* Sharon Yixuan Li is an Assistant Professor in the Department of Computer Sciences at the University of Wisconsin-Madison. Previously she was a postdoc research fellow in the Computer Science department at Stanford AI Lab. She completed her Ph.D. from Cornell University in 2017, where she was advised by John E. Hopcroft. She leads the organization of the ICML Workshop on Uncertainty and Robustness in Deep Learning in 2019 and 2020, and has served as area chair for NeurIPS, ICML, ICLR, and AAAI. Her broad research interests are in deep learning and machine learning. Her research develops algorithms and fundamental understandings to enable reliable open-world learning, which can function safely and adaptively in the presence of evolving and unpredictable data streams. She is the recipient of the Facebook Research Award, JPMorgan early-career faculty award, Madison Teaching and Learning Excellence fellowship, and was named Forbes 30Under30 in Science. Website: http://pages.cs.wisc.edu/~sharonli/ *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: March 9, 2022
by Mangannavar, Rajesh Devaraddi 03 Mar '22

03 Mar '22

Dear all, Our next AI seminar on *"Challenges and Opportunities in Out-of-distribution Detection" *by Sharon Yixuan Li is scheduled to be on March 9th, 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… *Challenges and Opportunities in Out-of-distribution Detection* Sharon Yixuan Li Assistant Professor Department of Computer Sciences University of Wisconsin-Madison *Abstract:* The real world is open and full of unknowns, presenting significant challenges for machine learning (ML) systems that must reliably handle diverse, and sometimes anomalous inputs. Out-of-distribution (OOD) uncertainty arises when a machine learning model sees a test-time input that differs from its training data, and thus should not be predicted by the model. As ML is used for more safety-critical domains, the ability to handle out-of-distribution data are central in building open-world learning systems. In this talk, I will talk about challenges, methods, and opportunities on uncovering the unknowns of deep neural networks for reliable predictions in an open world. *Speaker Bio:* Sharon Yixuan Li is an Assistant Professor in the Department of Computer Sciences at the University of Wisconsin-Madison. Previously she was a postdoc research fellow in the Computer Science department at Stanford AI Lab. She completed her Ph.D. from Cornell University in 2017, where she was advised by John E. Hopcroft. She leads the organization of the ICML Workshop on Uncertainty and Robustness in Deep Learning in 2019 and 2020, and has served as area chair for NeurIPS, ICML, ICLR, and AAAI. Her broad research interests are in deep learning and machine learning. Her research develops algorithms and fundamental understandings to enable reliable open-world learning, which can function safely and adaptively in the presence of evolving and unpredictable data streams. She is the recipient of the Facebook Research Award, JPMorgan early-career faculty award, Madison Teaching and Learning Excellence fellowship, and was named Forbes 30Under30 in Science. *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 Safety Reading Group
by Pope, Quintin 02 Mar '22

02 Mar '22

Hello everyone, We'll be meeting at 1 PM PST this Friday. We'll be discussing Quantifying Local Specialization in Deep Neural Networks <https://arxiv.org/abs/2110.08058>. Anyone interested in the internal structures/organization of neural networks should find this paper interesting. A neural network is locally specialized to the extent that parts of its > computational graph (i.e. structure) can be abstractly represented as > performing some comprehensible sub-task relevant to the overall task (i.e. > functionality). Are modern deep neural networks locally specialized? How > can this be quantified? In this paper, we consider the problem of taking a > neural network whose neurons are partitioned into clusters, and quantifying > how functionally specialized the clusters are. We propose two proxies for > this: importance, which reflects how crucial sets of neurons are to network > performance; and coherence, which reflects how consistently their neurons > associate with features of the inputs. To measure these proxies, we develop > a set of statistical methods based on techniques conventionally used to > interpret individual neurons. We apply the proxies to partitionings > generated by spectrally clustering a graph representation of the network's > neurons with edges determined either by network weights or correlations of > activations. We show that these partitionings, even ones based only on > weights (i.e. strictly from non-runtime analysis), reveal groups of neurons > that are important and coherent. These results suggest that graph-based > partitioning can reveal local specialization and that statistical methods > can be used to automatedly screen for sets of neurons that can be > understood abstractly. I look forward to meeting you then! 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

1 0

[CoRIS-List] Robotics Entrepreneurship
by Hurst, Jonathan 02 Mar '22

02 Mar '22

Hello All, Today at 11:30 AM I'll be doing a "TechCrunch Live" event, talking about how we pitched to investors and raised funding to start Agility Robotics, bringing the research we were doing at OSU to commercial deployments. For those of you interested in starting a business at some point, I hope this discussion could be useful to you. https://techcrunch.com/2022/03/02/agility-robotics-and-playground-global-jo… The link to registration is here: https://hopin.com/events/techcrunch-live-with-playground-global-and-agility… Jonathan ************************************************************ ***************** Jonathan W. Hurst Professor, Robotics School of Mechanical, Industrial, and Manufacturing Engineering Oregon State University 204 Rogers Hall Corvallis, OR 97331-6001 jonathan.hurst(a)oregonstate.edu Voice: (541) 737-7010 Fax: (541) 737-2600 http://mime.oregonstate.edu/research/drl/ ************************************************************ ******************

1 0

AI seminar: March 2, 2022
by Mangannavar, Rajesh Devaraddi 01 Mar '22

01 Mar '22

Dear all, Our next AI seminar on *"Considerations for More Scalable Trustworthy AI" *by Richard Mallah is scheduled to be on March 2nd (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… *Considerations for More Scalable Trustworthy AI* Richard Mallah Director of AI Projects The Future of Life Institute *Abstract:* As AI systems become less narrow in their capabilities and more general-purpose, new classes and levels of pitfalls present themselves, so techniques we bring to bear for the safety and ethical alignment of these systems will need to scale in new ways and require innovation increasingly on par with the sophistication of the primary learning functions of the system. As the differences between what a system can do and what it should do grow, desiderata we encounter even with narrower AI systems, like establishing safe bounds, practical interpretability, verification of key models or algorithms, minimizing negative side effects, maintaining operator control, and mitigation of unwanted biases each will need to account for levels of indirection previously unseen. In this talk, we attempt to explore the cultivation of a security mindset with respect to what more general systems can do wrong, and application of that to critical evaluation and design of safety techniques regarding their scalability or amenability to generality. It is notable, for instance, with increasing generality, that modeling context is increasingly important and that AI safety and AI ethics increasingly overlap. *Speaker Bio:* Richard Mallah is Director of AI Projects at The Future of Life Institute, where he does metaresearch, analysis, advocacy, strategy, and field building regarding technical, strategy, and policy aspects of transformative AI safety. From December 2015 Richard was on the founding team of the IEEE Global Initiative on Ethics of Autonomous and Intelligent Systems and continues to serve on its Executive Committee. He co-chairs the recurring SafeAI and AISafety technical safety workshops at AAAI and IJCAI, and in 2021 was the founding Executive Director of the Consortium on the Landscape of AI Safety, for which he was drafted because of his AI safety field landscaping & synthesis work at FLI. Richard has served in the Safety and the Labor & Economy Working Groups at Partnership on AI, has chaired the IEEE GIEAIS committees on AGI and on lethal autonomous weapons, and is an Honorary Senior Fellow at the Foresight Institute. Mr. Mallah has been working in machine learning and AI in industry for over twenty years, spanning many roles across R&D including algorithms research, research management, product team management, CTO, chief scientist, and strategy; in total he’s worked on over a hundred AI/ML-related technical projects from these different perspectives. Ever-focused on innovation yet mindful of managing risks, Richard has regularly aligned applied research drivers with novel research directions in trustworthy AI. In so doing, safety-related R&D he’s led has included: multiobjective GP agents with safety objectives, debiasing novel latent spaces, active-learning-enhanced automated ontology refactoring and alignment, explainability-enhanced conditional quasi-metric spaces, uncertainty-aware tight blends of symbolic and subsymbolic knowledge representations, liability-bearing-anomaly recognition, confidence-enhanced bayesian monte carlo analysis of operational risk, more robust multimodal LSTM systems, information-retrieval-enhanced transformer-based systems for more truthful NLG, and AI auditing methods. Richard advises AI safety startups, VC funds, incubators, academics, governments, international multistakeholder bodies, and NGOs on trustworthy AI, scalable AI safety, scalable AI ethics, wide-angle sustainability, ML model risk management, and assurance. Mr. Mallah has given dozens of invited talks globally on long-termist foresight, assurance, robustness, interpretability, and control of advanced AI and autonomous systems. He holds a degree in Intelligent Systems from Columbia University. This talk will be followed by a talk on the topic "*Challenges and Opportunities in Out-of-distribution Detection*" by professor Sharon Yixuan Li on 9th March (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 Alignment Reading Group
by Pope, Quintin 25 Feb '22

25 Feb '22

Hello everyone, We'll be meeting again at 1 PM PST on Friday. We'll discuss Locating and Editing Factual Knowledge in GPT <https://github.com/kmeng01/rome>. We investigate the mechanisms underlying factual knowledge recall in > autoregressive transformer language models. First, we develop a causal > intervention for identifying neuron activations capable of altering a > model's factual predictions. Within large GPT-style models, this reveals > two distinct sets of neurons that we hypothesize correspond to knowing an > abstract fact and saying a concrete word, respectively. This insight > inspires the development of ROME, a novel method for editing facts stored > in model weights. For evaluation, we assemble CounterFact, a dataset of > over twenty thousand counterfactuals and tools to facilitate sensitive > measurements of knowledge editing. Using CounterFact, we confirm the > distinction between saying and knowing neurons, and we find that ROME > achieves state-of-the-art performance in knowledge editing compared to > other methods. An interactive demo notebook, full code implementation, and > the dataset are available at this https URL <https://rome.baulab.info/>. I look forward to meeting you then! 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

2 1

[CoRIS-List] Today: OSU Robotics Seminar, Nick Gravish
by Ross L. Hatton 25 Feb '22

25 Feb '22

Hi All, As a reminder, our second winter Robotics Seminar will be *Today*, with a talk from Nick Gravish from the University of California, San Diego, who works in the area of bioinspired locomotion. The seminar will be from 10-11am, followed by an 11-11:30 student-only Q&A session. ** This seminar will be held in person in LINC 302. ** If you would like to meet with Nick, I have a few open meeting times. Please contact me if you would like one of them. --- A simulcast of the seminar will be available via Zoom https://oregonstate.zoom.us/j/92183247338?pwd=MENOWHgvSFVGYzNjZExDT2hWRUNxd… --- -Ross =========== 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 =========== Title: Design and control of emergent oscillations for flapping-wing flyers and swarming snakes Abstract: Locomotion in living systems and bio-inspired robots requires the generation and control of oscillatory motion. While a common method to generate motion is through modulation of time-dependent “clock” signals, in this talk we will motivate and study an alternative method of oscillatory generation through autonomous limit-cycle systems. Limit-cycle oscillators for robotics have many desirable properties including adaptive behaviors, entrainment between oscillators, and potential simplification of motion control. I will present two examples of the generation and control of autonomous oscillatory motion in bio-inspired robotics. First, I will describe our recent work to study the dynamics of wingbeat oscillations in “asynchronous” insects and how we can build these behaviors into micro-aerial vehicles. In the second part of this talk I will describe how simple snake-like robots with limit-cycle gaits enable swarms to synchronize their movement through contact and without communication. More broadly in this talk I hope to motivate why we should look to autonomous dynamical systems for designing and controlling emergent locomotor behaviors in robotics. Bio: Dr. Nick Gravish received his PhD from Georgia Tech where he used robots as physical models to motivate and study aspects of biological locomotion. During his post-doc Gravish worked in the microrobotics lab of Rob Wood at Harvard, where he gained expertise in designing and studying insect-scale robots. Gravish is currently an assistant professor at UC San Diego in the Mechanical and Aerospace Engineering department. His lab bridges the gap between bio-inspiration, biomechanics, and robotics, towards the development of new bio-inspired robotic technologies to improve the adaptability and resilience of mobile robots.

1 0