AAAI Technical Track on Machine Learning I, Boston Children's Hospital, Harvard Medical School, USA InProceedings of the AAAI Conference on Artificial Intelligence (Vol 35,), Filan, D.*, Casper, S.*, Hod, S.*, Wild, C., Critch, A., & Russell, S. (2021). I like historic buildings. Casper, S.*, Hod, S.*, Filan, D.*, Wild, C., Critch, A., & Russell, S. (2021). Open Peer Review. Next year, Casper hopes hell be in graduate school, but if that doesnt work out, hell take a machine learning engineering job for a few years. arXiv, Casper, S. (2020). He pasted playing cards, magazine clippings, and postcards on the walls of his dorm room. He pasted playing cards, magazine clippings, and postcards on the walls of his dorm room. How do you usually write your name as author of a paper? History of Medicine in Europe and North America; Neuroscience and Society; Introduction to Culture, Biology and Society; European History, the long 19th Century; Science, Technology and Society, The history of traumatic brain injuries and concussions; Physiology in the Anglo-American World in modernity; The politics of neuroscience; the History of Disturbances of Consciousness; the Patient in Medical History, Casper, S. (2014). Every week on Wed, until Nov 4, 2020, 4 occurrence(s) For the past few months, hes been exploring witchy YouTube channels and podcasts and occasionally pretending to be a witch on Reddit threads. Once, he took a book of poetry from the Pforzheimer Library and cut out a bunch of poems. His answer started with what I expected his roommates and the Effective Altruism Club but then he thought for a moment. Someone in my hall was like, You better find that lizard. We gratefully acknowledge the support of the OpenReview Sponsors. Also add any other names you have authored papers under.
80 Nott Terrace, Schenectady, NY 12308 518-631-9831 (Optional). Manchester: Manchester University Press. Lately, I have been particularly interested in finding (mostly) automated ways of finding/fixing flaws in how deep neural networks handle human-interpretable concepts. 199 Dennings Ave., Beacon, New York 12508 845-838-1600 Open Access. The History of the Brain and Mind Sciences: Technique, Technology and Therapy.Rochester NY: University of Rochester Press. Login to access subscriber-only resources. Nov 4, 2020 12:00 PM, July 24, 2022 - July 27, 2022 Oct 14, 2020 12:00 PM
Enter your feedback below and we'll get back to you as soon as possible. The following articles are merged in Scholar. See also announcement of Hoopes Prize for Stephen Casper. Topic: KLAB lab meeting Feel free to use it to send me anonymous, constructive feedback about how I can be better. +1 207-244-3600, Check-in: Jul 24, 2022 12:00am
Oct 28, 2020 12:00 PM I've had first hand experience in marine biology as well. The system can't perform the operation now. CBMM is committed to developing and supporting new interdisciplinary collaborations that will further our understanding of intelligence. You can also ask me about my hissing cockroaches, a bracelet I made for a monkey, witchcraft, what I learned from getting my genome sequenced, a time I helped with a mammoth undertaking, a necklace that I wear every full moon, or a jar I keep in my windowsill. Back at school, Casper kept a jar of buttons, paper clips, screws, pins, and pieces of washers. Dr Gabriel Kreiman is inviting you to a scheduled Zoom meeting. 132-143). Frivolous Units: Wider Networks Are Not Really That Wide. Theres one witchcraft practice that particularly intrigues Casper: making an altar. I love archives and libraries. I also have a personal feedback form. Im working on a few projects involving adversarial policies in reinforcement learning, learning rewards from humans, interpretable adversarial examples, and a few other ideas.
Im interested in it ethnographically, but I also think that hanging around witches in Boston just seems fun as hell, right? he says. He loves the meaninglessness of the little things he acquires. Ill miss the Adams tunnels, he remarks. I really hate lizards, and I was like, No, I just need a lizard.. To submit a bug report or feature request, you can use the official OpenReview GitHub repository:Report an issue. The Neurological Patient in History. Over the summer in 2018, Casper was working on a bioinformatics project under the Church Lab and living in Winthrop House when two of his cockroaches mated. Take a Virtual Tour, Organizational Chart Enter your education and career history. Enter email addresses associated with all of your current and historical institutional affiliations, as well as all your previous publications, and the Toronto Paper Matching System. The institution domain is used for conflict of interest detection and institution ranking. For now, Im not posting my resume/CV here, but please email me if youd like to talk about projects or opportunities. It has been adopted by over 30 Effective Altruism groups in the US, Canada, and China and is approaching over a thousand alumni. Time: Oct 14, 2020 12:00 PM Eastern Time (US and Canada) Center for Brains, Minds, and Machines (CBMM) 
Achilles Heels for AGI/ASI via Decision Theoretic Adversaries. Saleh, A., Deutsch, T., Casper, S., Belinkov, Y., & Shieber, S. M. (2020, July). OpenReview is a long-term project to advance science through improved peer review, with legal nonprofit status through Code for Science & Society. Im willing to be a walking commercial for the EA Club, Casper says with a laugh, emphasizing that his involvement in the movement wont stop after graduation. 
35 No. Thats why I like to paste shit all over my room. His trinket collection feeds a curiosity that exists just for the sake of itself, the same kind of pure fascination that lands someone in a coven meeting or with dozens of baby cockroaches.
Spared the circle of life stuff, Trevor now enjoys his temporary home in Idaho Falls, Idaho, where Casper grew up. *, Casper, S.*, Liu, S., Yang, K., & Church, G. (2021). Clusterability in Neural Networks. The Neurologists: A History of a Medical Specialty in Modern Britain, c. 1789-2000. Open Publishing. Try again later. Switching to virtual learning wasnt too difficult, he tells me. Hod, S.*, Casper, S.*, Filan, D.*, Wild, C., Critch, A., & Russell, S. (2021). Center for Brains, Minds, and Machines (CBMM), Boston Children's Hospital, Harvard Medical School Casper, S.*, Boix, X. Graphical Clusterability and Local Specialization in Deep Neural Networks, Pair^2Struct Workshop, ICLR 2022. Despite the species name, Trevor stays quiet hes pretty docile in the wintertime, explains Casper, who has kept pet cockroaches in his dorm room since he was a freshman. Hi, Im Stephen Casper, but most people call me Cas. But Caspers propensity for collecting discarded items also speaks to his ability to notice them in the first place. The Arete Fellowship, a program I founded with the Harvard Effective Altruism club. Add links to your profiles on other sites. PKP Publishing Services Network, Copyright 2019, Association for the Advancement of Artificial Intelligence. Its for the same reason that a crow loves a piece of tinfoil: because its just a thing thats interesting-looking, he says. Frivolous Units: Wider Networks Are Not Really That Wide, Achilles Heels for AGI/ASI via Decision Theoretic Adversaries, Probing Neural Dialog Models for Conversational Understanding. The ones marked, S Casper, X Boix, V D'Amario, L Guo, M Schrimpf, K Vinken, G Kreiman, Proceedings of the AAAI Conference on Artificial Intelligence 35 (8), 6921-6929, D Filan, S Casper, S Hod, C Wild, A Critch, S Russell, A Saleh, T Deutsch, S Casper, Y Belinkov, S Shieber, Y Chen, E Hysolli, A Chen, S Casper, S Liu, K Yang, C Liu, G Church, S Hod, S Casper, D Filan, C Wild, A Critch, S Russell, New articles related to this author's research, Professor, Children's Hospital, Harvard Medical School, Frivolous units: Wider networks are not really that wide, Probing neural dialog models for conversational understanding, Multiplex base editing to convert TAG into TAA codons in the human genome, One thing to fool them all: Generating interpretable, universal, and physically-realizable adversarial features, Detecting Modularity in Deep Neural Networks, Achilles Heels for AGI/ASI via Decision Theoretic Adversaries, Eicient and Insidious Adversaries in Deep Reinforcement Learning. 8: AAAI-21 Technical Tracks 8. Stephen Casper. Casper studies applications of machine learning involving AI safety. Chen, Y. *, DAmario, V., Guo, L., Schrimpf, M., Vinken, K., & Kreiman, G. (2021). My main focus is in developing tools for more interpretable and robust AI. Enter all advisors, co-workers, and other people that should be included when detecting conflicts of interest. Frivolous Units: Wider Networks Are Not Really That Wide. 2022 by Clarkson University | Contact the Webmaster, Beacon Institute for Rivers and Estuaries, Early College Program: The Clarkson School, Earl R. and Barbara D. Lewis School of Health Sciences, Clarkson Center for Complex Systems Science (C3S2), The Center for Advanced Materials Processing (CAMP), Center for Air and Aquatic Resources Engineering & Science (CAARES), Center for Identification Technology Research (CITeR), The Center for Rehabilitation Engineering, Science & Technology (CREST), NYS Center of Excellence in Healthy Water Solutions, CARES Act Reports & Consumer Information Disclosures (HEOA). For each line, enter comma-separated keyphrases representing an intersection of your interests.
Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation. Even though hes not superstitious, he loves the concept of arranging witchy things on a small table. Ill miss walking past that vent that blows warm air next to Canaday in the winter.. arXiv. Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology, USA, Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology, USA, Neuroscience Graduate Program, University of California San Francisco, USA, Center for Brains, Minds, and Machines (CBMM) In the long term, he wants to pursue a career publishing papers involving machine learning, either in academia or for a nonprofit organization. A statistics concentrator whos for all intents and purposes a [computer science] person, Caspers academic tasks mostly consist of coding and doing math. Formerly, I have worked with the Harvard Kreiman Lab (where I did my undergrad) and the Center for Human-Compatible AI. Casper, S., Jacyna, L. S. (2012), eds. andGavrus G. (in press), eds. *, Hysolli, E.*, Chen, A. Efficient and insidious adversaries in deep reinforcement learning. Seawall Campground, Southwest Harbor, Seawall Campground
Staff writer Josie F. Abugov can be reached at josie.abugov@thecrimson.com. Robust Feature-Level Adversaries are Interpretability Tools. I like meeting and talking with new people. Graphical Clusterability and Local Specialization in Deep Neural Networks, Detecting Modularity in Deep Neural Networks, Robust Feature-Level Adversaries are Interpretability Tools. Harvard University (2021). Think of each line as a query for papers in which you would have expertise and interest. I've loved history and science my entire life. Back at school, he kept a jar of buttons, paper clips, screws, pins, and pieces of washers. This quirky addition to such a professional page might have something to do with Caspers dedication to effective altruism, the idea that we should employ evidence in philanthropy to build a better world, according to the Harvard College Effective Altruism Club website. arXiv, Casper, S.*, Nadeau, M.*, Hadfield-Menell, D. Kreiman, G (2021). This information is crucial for deduplicating users, and ensuring you see your reviewing assignments. Oct 21, 2020 12:00 PM
I was born in a small town in Minnesota. CasperS. I'd prefer not to own a mobile phone. Overwhelmed with 30 newborn cockroaches, Casper wanted to do some circle of life stuff and feed them to a lizard, so he took to the Pforzheimer House mailing list and asked if anyone could temporarily loan him a lizard. bioRxiv. Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology, USA, Representation Learning, Bio-inspired Learning, Learning on the Edge & Model Compression, Proceedings of the AAAI Conference on Artificial Intelligence, Vol. But I have a twitter account. 
When hes not parenting cockroaches, preventing the AI apocalypse, or becoming a more effective person, Casper spends his spare time learning about witchcraft. Multiplex base editing to convert TAG into TAA codons in the human genome. Next year, Casper hopes hell be in graduate school, but if that doesnt work out, hell take a machine learning engineering job for a few years. Rochester NY: University of Rochester Press, 8 Clarkson Ave., Potsdam, New York 13699 315-268-6400 Probing Neural Dialog Models for Conversational Understanding. Some of his projects on neural networks and adversarial policies can be found on his website, stephencasper.com, below an introduction that includes a form for anonymous, constructive feedback about how I can be a better or more effective person.. Between paganism and Wicca and just independent people practicing witchcraft, there are lots of witches who just do many different things, he says, as I scour my Etsy purchase history to send him a link to an obsidian crystal necklace I bought. Feel free to reach out. He also plans on working within a subgroup of the machine learning research community on preventing the artificial intelligence apocalypse, which he says hes totally into, but not in a creepy death cult way., At the moment, Casper isnt too worried about his thesis or the apocalypse hes been writing deep learning papers and blog posts for the past few years. Part of the Check-out: Jul 27, 2022 12:00am, Understanding the differences between biological and computer vision, Giorgia Dellaferrera gives a talk at ICML 2022, Interview with Prof. Kreiman about research and career paths, Annual meeting of the Cogitate Consortium, 2022, Global neuronal workspace versus Integrated information theory.
Theyll wear eyeliner and paint their faces, he tells me without further explanation, and join one of the open-invite covens that theyve already located through Facebook groups. Im a second year Ph.D student at MIT in Computer Science (EECS) in the Algorithmic Alignment Group advised by Dylan Hadfield-Menell. arXiv. I've worked in neurology and neuroscience laboratories. Their, This "Cited by" count includes citations to the following articles in Scholar. For example: deep learning, RNNs, dependency parsing. 2020 Center for Brain, Minds, & Machines, Modeling Human Goal Inference as Inverse Planning in Real Scenes, Computational models of human social interaction perception, Invariance in Visual Cortex Neurons as Defined Through Deep Generative Networks, Sleep Network Dynamics Underlying Flexible Memory Consolidation and Learning, Neurally-plausible mental-state recognition from observable actions, Undergraduate Summer Research Internships in Neuroscience, Shared Visual Representations in Human & Machine Intelligence (SVRHM) 2021, Shared Visual Representations in Human & Machine Intelligence (SVRHM) 2020, REGML 2020 | Regularization Methods for Machine Learning, MLCC 2020 @ simula Machine Learning Crash Course, Shared Visual Representations in Human and Machine Intelligence (SVRHM) Workshop 2019, A workshop on language and vision at CVPR 2019, A workshop on language and vision at CVPR 2018, Learning Disentangled Representations: from Perception to Control, A workshop on language and vision at CVPR 2017, Science of Intelligence: Computational Principles of Natural and Artificial Intelligence, CBMM Workshop on Speech Representation, Perception and Recognition, Deep Learning: Theory, Algorithms and Applications, Biophysical principles of brain oscillations and their meaning for information processing, Neural Information Processing Systems (NIPS) 2015, Engineering and Reverse Engineering Reinforcement Learning, Learning Data Representation: Hierarchies and Invariance. Plus, he thinks an altar would complement his goblin-y aesthetic and fuel his habit of collecting little objects. InProceedings of the 2nd Workshop on Natural Language Processing for Conversational AI(pp. 
Casper studies applications of machine learning involving AI safety, and his senior thesis focuses on a type of threat toward reinforcement learning systems that could be relevant for self-driving cars. Casper, S., Boix, X., DAmario, V., Guo, L., Schrimpf, M., Vinken, K., & Kreiman, G. (2021). For ongoing positions, leave the end field blank. Youre welcome to email me (and do it a second time if I dont respond). Stephen M. Casper 21 smiles for a Zoom screenshot while Trevor, his Madagascar hissing cockroach, crawls across his left cheek. Im also an Effective Altruist trying to do the most good I can. Detecting Modularity in Deep Neural Networks. COVID-19 regulations permitting, Casper is itching to attend coven meetings in Boston with a few of his roommates in the spring. Multiplex base editing to convert TAG into TAA codons in the human genome. You can find me on Google Scholar, Github, LinkedIn, or LessWrong. But I think I typed that email in a pretty ambiguous way because people started asking me if I found the lizard, Casper recalls. As a board member of the club, Casper founded, chaired, and designed curriculum for the Arete Fellowship program, which over 25 Effective Altruism university groups have since adopted. At the end of our conversation, I ask him what hes going to miss most about Harvard. Southwest Harbor Research interests of mine include interpretability, adversaries, robust reinforcement learning, and decision theory.