Meta tags:

Headings (most frequently used words):

and, robot, my, to, human, learning, ai, in, the, teaching, model, university, policy, specify, inspect, design, methods, about, statements, from, of, reward, through, bayes, trex, sampling, by, hi, serena, cs, prof, at, brown, office, is, cit, 427, was, aaas, fellow, senate, sidequests, media, coverage, advocacy, conferences, journals, workshops, people, must, be, easily, able, revise, behaviors, tools, enable, these, interactions, you, can, read, more, interests, research, diversity, 2023, faculty, job, search, csci, 1952a, interaction, perils, trial, error, misdesign, overfitting, invalid, task, specifications, optimal, advantage, preferences, mistaking, it, for, bayesian, approach, transparency, example, rocus, controller, understanding, via, an, extension, do, feature, attribution, work, communicating, logical, effectively, revisiting, lens, concept, theory, varying, how, we, teach, adding, contrast, helps, humans, learn, motions, machine, practice, outside, big, tech, resource, constraints, challenge, responsible, development, piggybacking, robots, overtrust, dormitory, security, phd, comics, soonish, kelly, zach, weinersmith, science, equity, inclusion,

Text of the page (most frequently used words):
and (47), learning (33), #serena (26), robot (26), human (26), booth (25), the (25), conference (24), webpage (21), for (18), aaai (15), shah (15), how (14), hri (14), julie (13), intelligence (13), 2021 (13), reward (13), artificial (12), robots (11), 2022 (11), project (11), published (11), design (10), from (10), about (9), 2017 (9), piggybacking (9), workshop (8), preferences (8), interaction (8), you (8), humans (7), methods (7), 2023 (7), bradley (7), knox (7), international (7), machine (7), specifications (7), these (7), that (7), yilun (6), zhou (6), sampling (6), scott (6), niekum (6), peter (6), stone (6), acm (6), ethics (6), model (6), teaching (6), reinforcement (6), science (6), mit (6), policy (6), work (6), 2020 (5), concept (5), attribution (5), ieee (5), bayes (5), trex (5), 2024 (5), with (5), this (5), behaviors (5), understand (4), stephane (4), hatgis (4), kessell (4), logical (4), through (4), alessandro (4), allievi (4), trial (4), error (4), 2025 (4), study (4), soonish (4), phd (4), more (4), systems (4), their (4), are (4), find (4), can (4), better (4), via (3), examples (3), neural (3), approach (3), elena (3), glassman (3), models (3), motions (3), preference (3), optimal (3), advantage (3), mistaking (3), university (3), security (3), 2019 (3), statements (3), practice (3), tech (3), resource (3), responsible (3), development (3), controller (3), understanding (3), transparency (3), feature (3), theory (3), making (3), overfitting (3), functions (3), misdesign (3), research (3), specification (3), president (3), using (3), trust (3), news (3), help (3), computing (3), whether (3), should (3), people (3), email (2), intent (2), christian (2), muise (2), knowledge (2), compilation (2), ankit (2), sanjana (2), sharma (2), sarah (2), chung (2), your (2), marco (2), ribeiro (2), priority (2), correctly (2), attribute (2), interactive (2), varying (2), teach (2), adding (2), contrast (2), helps (2), learn (2), icml (2), sigurdur (2), orn (2), adalgeirsson (2), anca (2), dragan (2), overtrust (2), dormitory (2), joint (2), ijcai (2), communicating (2), effectively (2), society (2), aies (2), outside (2), big (2), constraints (2), challenge (2), corl (2), rocus (2), bayesian (2), example (2), revisiting (2), lens (2), multidisciplinary (2), decision (2), extended (2), abstract (2), graduate (2), perils (2), invalid (2), task (2), matthew (2), diversity (2), tmlr (2), rlc (2), towards (2), practitioners (2), strong (2), advocate (2), inclusion (2), women (2), served (2), pictured (2), above (2), vice (2), inform (2), not (2), advocacy (2), friday (2), harvard (2), seas (2), cookie (2), look (2), like (2), kelly (2), zach (2), weinersmith (2), comics (2), alumni (2), profile (2), evaluation (2), media (2), which (2), non (2), consequences (2), some (2), representative (2), looking (2), expressive (2), writing (2), inspect (2), arise (2), regret (2), write (2), experts (2), specify (2), csci (2), 1952a (2), brown (2), html5, github, twitter, plan, activity, recognition, pair, salomon, wollenstein, betech, yasaman, khazaeni, modeling, blackbox, agent, behaviour, statistical, relational, starai, prediction, matching, networks, probabilistic, programming, icra, social, sihr, space, informed, neurips, xai4debugging, priorities, naacl, trustworthy, natural, language, processing, trustnlp, yiming, zheng, irrationality, rationale, tiffany, horter, many, facets, based, mfpl, workshops, james, tompkin, hanspeter, pfister, radhika, nagpal, krzysztof, gajos, jim, waldo, evaluating, interpretability, map, aspen, hopkins, nadia, figueroa, features, spotlight, partial, return, poorly, explains, student, descent, considered, harmful, proposal, studying, allen, chang, fontaine, stefanos, nikolaidis, maja, matarić, quality, generative, synthetic, data, transactions, septia, rani, sarath, sreedharan, goals, rewards, comparative, objective, mechanisms, calarina, muslimani, kerrick, johnstonbaugh, suyog, chandramouli, taylor, improving, alignment, metric, position, consumer, protection, inalienable, defense, safety, united, states, conferences, journals, underrepresented, minorities, high, school, students, introductory, class, taught, puebla, mexico, gw6, course, equity, make, just, equitable, colleague, willie, boag, endorsement, senator, alexander, initiative, spectrum, video, automation, motherboard, would, gave, thing, love, janelle, shane, offers, excerpt, crossover, 2016, into, well, explanation, new, technology, think, critically, ocw, podcast, embed, education, coverage, award, winning, undergraduate, senior, thesis, set, out, answer, question, place, too, much, robotic, specifically, physical, domain, interviewed, industry, startups, government, companies, use, developing, products, analyze, interviews, thematic, analysis, sidequests, preliminary, test, applying, insights, variation, assist, cognitive, theories, interfaces, especially, tasks, come, behavioral, patterns, encoded, generally, maintain, mitigate, uncertainty, beliefs, best, present, sentences, different, forms, easier, harder, parse, resilient, than, anticipated, principled, mechanism, assessing, contribute, growing, body, literature, suggesting, cannot, trusted, wild, sample, show, exposing, revision, dynamical, system, extension, network, tool, after, algorithm, optimize, person, assess, has, learned, behavior, meets, needs, expectations, aligned, typical, rlhf, approaches, assume, only, trajectory, segments, sums, past, showed, consider, instead, supported, unsanctioned, but, implications, widespread, have, been, studied, conduct, empirical, computational, user, experiments, leads, overfit, otherwise, misdesigned, even, trivial, setting, function, rampant, critical, yet, notoriously, hard, because, lack, common, sense, reasoning, easy, result, unintended, potentially, dangerous, side, effects, interpreted, fall, website, external, want, apply, mention, name, application, postdoc, interest, program, must, easily, able, tools, enable, interactions, read, interests, faculty, job, search, revise, prof, office, cit, 427, was, aaas, fellow, senate, publications, resume,

Text of the page (random words):
serena booth resume publications advocacy media research about hi i m serena i m a cs prof at brown university my office is cit 427 i was a aaas ai policy fellow in the u s senate people must be easily able to specify model inspect and revise ai and robot behaviors i design methods and tools to enable these interactions you can read more about my interests in my research teaching and diversity statements from my 2023 faculty job search if you are external to brown and want to do a phd with me apply to the cs phd program and mention my name in your application if you d like to do a postdoc email me i m looking for interest in human ai human robot interaction reinforcement learning and or ai policy teaching csci 1952a human ai interaction fall 2025 csci 1952a human ai interaction website specify writing specifications for ai systems is critical yet notoriously hard because these systems lack common sense reasoning making it easy to write specifications that result in unintended and potentially dangerous side effects i study how experts and non experts write specifications and how these specifications should be interpreted the perils of trial and error reward design misdesign through overfitting and invalid task specifications trial and error reward design is unsanctioned but the implications of this widespread practice have not been studied we conduct empirical computational and user study experiments and we find that trial and error leads to the design of reward functions which are overfit and otherwise misdesigned even in a trivial setting we find that reward function misdesign is rampant published at aaai conference on artificial intelligence 2023 project webpage learning optimal advantage from preferences and mistaking it for reward typical reinforcement learning from human preferences rlhf approaches assume that human preferences arise only from trajectory segments sums of reward in past work we showed that regret is a better model of human preferences published at tmlr 2024 in this work we consider the consequences if preferences arise instead from this better supported regret preference model published at aaai conference on artificial intelligence 2024 project webpage inspect after writing a specification and using some algorithm to optimize it how can a person assess whether a robot or an ai has learned the behavior that meets their needs and expectations is it aligned to their intent bayes trex a bayesian sampling approach to model transparency by example looking at expressive examples can help us better understand neural network behaviors and design better models bayes trex is a tool to find these expressive examples published at aaai conference on artificial intelligence 2021 project webpage rocus robot controller understanding via sampling an extension to bayes trex we sample representative robot behaviors we show how exposing these representative behaviors can help with the revision of a dynamical system robot controller s specifications published at conference on robot learning corl 2021 project webpage do feature attribution methods work we design a principled evaluation mechanism for assessing priority attribution methods and contribute to the growing body of literature suggesting these methods cannot be trusted in the wild published at aaai conference on artificial intelligence 2022 project webpage communicating logical statements effectively how should we best present logical sentences to a human i study whether different logical forms are easier or harder for people to parse i find that people are more resilient than anticipated published at international joint conference on ai ijcai 2019 project webpage model how do humans come to understand the behavioral patterns encoded in a specification more generally how do humans maintain and mitigate uncertainty about their beliefs about ai systems revisiting human robot teaching and learning through the lens of human concept learning theory we look at how cognitive theories of human concept learning should inform human robot interaction interfaces especially for teaching and learning tasks published at acm ieee international conference on human robot interaction hri 2022 project webpage varying how we teach adding contrast helps humans learn about robot motions in this preliminary work we test the consequences of applying some of the insights from the variation theory of learning to assist humans in learning about robot motions published at hri workshop on human interactive robot learning 2023 project webpage sidequests machine learning practice outside big tech resource constraints challenge responsible development we interviewed industry practitioners from startups government and non tech companies about their use of machine learning in developing products we analyze these interviews with thematic analysis published at aaai acm conference on artificial intelligence ethics and society aies 2021 project webpage piggybacking robots human robot overtrust in university dormitory security my award winning undergraduate senior thesis a project which set out to answer the question of whether we place too much trust in robotic systems specifically in the physical security domain published at acm ieee international conference on human robot interaction hri 2017 project webpage media coverage phd comics soonish by kelly and zach weinersmith ethics in computing an ocw podcast on my work to embed ethics in cs education at mit ethics in computing learning to think critically about machine learning ethics in computing a new resource for teaching responsible technology development hri 2022 human concept learning mit news how to help humans understand robots aaai 2022 feature attribution evaluation mit news how well do explanation methods work aaai 2021 bayes trex mit news more transparency and understanding into machine behaviors alumni profile harvard seas alumni profile on serena booth a b 2016 hri 2017 piggybacking robots phd comics soonish crossover hri 2017 piggybacking robots soonish by kelly and zach weinersmith hri 2017 piggybacking robots science friday if a robot offers you a cookie soonish excerpt hri 2017 piggybacking robots you look like a thing and i love you by janelle shane hri 2017 piggybacking robots motherboard vice would you trust this robot if it gave you a cookie hri 2017 piggybacking robots harvard seas in automation we trust hri 2017 piggybacking robots ieee spectrum video friday advocacy science policy in 2021 2022 i served as president and in 2020 2021 as vice president of mit s science policy initiative i advocate for using science to inform policy and for using policy to make science just and equitable pictured above with colleague willie boag not an endorsement for senator alexander equity and inclusion i m a strong advocate for the inclusion of women and underrepresented minorities in science in 2019 i served as co president of mit s gw6 graduate women of course 6 pictured above high school students from an introductory cs class i taught in puebla mexico conferences and journals position strong consumer protection is an inalienable defense for ai safety in the united states serena booth international conference on machine learning icml 2025 towards improving reward design in rl a reward alignment metric for rl practitioners calarina muslimani kerrick johnstonbaugh suyog chandramouli serena booth w bradley knox matthew e taylor reinforcement learning conference rlc 2025 goals vs rewards towards a comparative study of objective specification mechanisms septia rani serena booth sarath sreedharan reinforcement learning conference rlc 2025 models of human preference for learning reward functions w bradley knox stephane hatgis kessell serena booth scott niekum peter stone alessandro allievi transactions on machine learning research tmlr 2024 webpage learning optimal advantage from preferences and mistaking it for reward w bradley knox stephane hatgis kessell sigurdur orn adalgeirsson serena booth anca dragan peter stone scott niekum aaai conference on artificial intelligence 2024 webpage quality diversity generative sampling for learning with synthetic data allen chang matthew fontaine serena booth maja matarić stefanos nikolaidis aaai conference on artificial intelligence 2024 the perils of trial and error reward design misdesign through overfitting and invalid task specifications serena booth w bradley knox julie shah scott niekum peter stone alessandro allievi aaai conference on artificial intelligence 2023 webpage extended abstract graduate student descent considered harmful a proposal for studying overfitting in reward functions serena booth w bradley knox julie shah scott niekum peter stone alessandro allievi multidisciplinary conference on reinforcement learning and decision making 2022 spotlight extended abstract partial return poorly explains human preferences w bradley knox stephane hatgis kessell serena booth scott niekum peter stone alessandro allievi multidisciplinary conference on reinforcement learning and decision making 2022 revisiting human robot teaching and learning through the lens of human concept learning theory serena booth sanjana sharma sarah chung julie shah elena l glassman acm ieee international conference on human robot interaction hri 2022 webpage do feature attribution methods correctly attribute features yilun zhou serena booth marco ribeiro julie shah aaai conference on artificial intelligence 2022 webpage bayes trex a bayesian sampling approach to model transparency by example serena booth yilun zhou ankit shah julie shah aaai conference on artificial intelligence 2021 webpage rocus robot controller understanding via sampling yilun zhou serena booth nadia figueroa julie shah conference on robot learning corl 2021 webpage machine learning practice outside big tech how resource constraints challenge responsible development aspen hopkins serena booth aaai acm conference on artificial intelligence ethics and society aies 2021 webpage evaluating the interpretability of the knowledge compilation map communicating logical statements effectively serena booth christian muise julie shah international joint conference on ai ijcai 2019 webpage piggybacking robots human robot overtrust in university dormitory security serena booth james tompkin hanspeter pfister jim waldo krzysztof gajos radhika nagpal acm ieee international conference on human robot interaction hri 2017 webpage workshops learning optimal advantage from preferences and mistaking it for reward w bradley knox stephane hatgis kessell sigurdur orn adalgeirsson serena booth anca dragan peter stone scott niekum 2023 icml workshop on the many facets of preference based learning mfpl 2023 varying how we teach adding contrast helps humans learn about robot motions tiffany horter elena l glassman julie shah serena booth hri workshop on human interactive robot learning 2023 webpage the irrationality of neural rationale models yiming zheng serena booth julie shah yilun zhou 2022 naacl workshop on trustworthy natural language processing trustnlp 2022 do priority attribution methods correctly attribute priorities yilun zhou serena booth marco ribeiro julie shah neurips 2021 xai4debugging workshop 2021 how to understand your robot a design space informed by human concept learning serena booth sanjana sharma sarah chung julie shah elena l glassman icra 2021 workshop on social intelligence in humans and robots sihr 2021 sampling prediction matching examples in neural networks a probabilistic programming approach serena booth ankit shah yilun zhou julie shah aaai 2020 workshop on statistical relational artificial intelligence starai 2020 modeling blackbox agent behaviour via knowledge compilation christian muise salomon wollenstein betech serena booth julie shah yasaman khazaeni aaai 2020 workshop on plan activity and intent recognition pair 2020 twitter github email design html5 up