Meta tags:

Headings (most frequently used words):

and, machine, learning, student, response, systems, musings, about, other, things, posts, search, this, blog, constrained, mdps, the, reward, hypothesis, numerical, errors, perturbation, analysis, year, later, archive, labels, report, abuse,

Text of the page (most frequently used words):
the (47), that (17), this (14), and (11), #learning (10), are (10), more (9), systems (7), should (7), #machine (6), student (6), you (6), other (6), socrative (6), show (5), analysis (5), there (5), which (5), what (5), like (5), device (5), not (5), class (5), apr (4), read (4), have (4), first (4), use (4), our (4), they (4), apps (4), email (4), pinterest (4), facebook (4), get (4), link (4), some (4), was (4), numerical (4), errors (4), about (4), blog (4), reward (3), mdps (3), constrained (3), sept (3), post (3), response (3), out (3), students (3), can (3), with (3), back (3), computers (3), will (3), september (3), 2013 (3), used (3), material (3), fact (3), less (2), tools (2), statistics (2), hypothesis (2), perturbation (2), optimization (2), models (2), mathematics (2), markovian (2), decision (2), theory (2), bayesian (2), may (2), mar (2), visit (2), posts (2), comment (2), one (2), fall (2), 2012 (2), few (2), iclicker (2), iclickers (2), overall (2), cost (2), but (2), when (2), don (2), for (2), comments (2), experience (2), after (2), semester (2), ago (2), here (2), see (2), kind (2), had (2), try (2), everyone (2), new (2), whether (2), would (2), think (2), error (2), set (2), nice (2), today (2), however (2), issue (2), well (2), recent (2), want (2), cmdps (2), multiple (2), while (2), musings (2), things (2), report, abuse, x11, thunderbird, technology, technical, svn, supervised, stability, science, sample, complexity, representation, reinforcement, presentation, powerpoint, pdf, optogenetics, occam, razor, non, parametric, model, selection, mixing, matlab, processes, mac, osx, least, squares, latex, keynote, jbig2, image, processing, hilbert, matrix, galerkin, method, frequentist, approach, exploration, djvu, curse, dimensionality, conditioning, number, compression, clinical, trials, blogging, beamer, style, artificial, intelligence, approximation, annoyance, aggregation, labels, jul, aug, oct, nov, archive, profile, csaba, szepesvári, theme, images, michael, elkan, powered, blogger, plenty, srss, choose, brief, document, summarizes, found, web, before, radio, communication, based, differ, characteristics, type, devices, classical, require, buy, communicates, receiver, teacher, possess, since, school, uses, let, focus, them, generation, assuming, sells, bookstore, major, who, likes, pay, limitation, bound, smart, classrooms, thus, your, own, computer, projection, need, switch, between, screens, results, poll, makes, icli, promised, come, though, end, passed, long, thoughts, very, positive, during, week, polled, how, large, percentage, has, wifi, enabled, could, were, willing, bring, decided, give, helped, tremendously, stay, top, knows, way, follows, every, block, usually, slides, inserted, questions, verify, got, all, extra, work, put, designing, due, done, something, similar, without, anyways, did, feel, much, difference, once, year, later, hates, love, machines, infinite, precision, really, hated, sounded, subject, study, annoying, side, effect, imperfect, boring, detail, miles, away, from, anything, anyone, ever, consider, part, convince, opposite, true, even, ideas, lessons, learned, asks, question, doing, care, understood, appreciated, received, attention, doubt, discussed, any, textbooks, beyond, usual, caveat, beware, scary, illustrate, questi, been, looong, posted, mean, dead, slow, steady, wins, race, right, anyhow, write, process, prompted, eugene, feinberg, pioneer, department, also, growing, interest, cmpds, community, paper, impatient, readers, cmdp, mdp, except, functions, objective, others, restrict, policies, now, seems, often, than, problems, solve, easiest, specify, using, objectives, borderline, tautology, example, given, current, sad, situation, hard, escape, deciding, interventions, government, apply, limit, spread, virus, maintaining, economic, march, 2020, home, search, skip, main, content,

Text of the page (random words):
musings about machine learning and other things skip to main content search this blog musings about machine learning and other things home more posts constrained mdps and the reward hypothesis get link facebook x pinterest email other apps march 20 2020 it s been a looong ago that i posted on this blog but this should not mean the blog is dead slow and steady wins the race right anyhow i am back and today i want to write about constrained markovian decision process cmdps the post is prompted by a recent visit of eugene feinberg a pioneer of cmdps of our department and also by a growing interest in cmpds in the rl community see this this or this paper for impatient readers a cmdp is like an mdp except that there are multiple reward functions one of which is used to set the optimization objective while the others are used to restrict what policies can do now it seems to me that more often than not the problems we want to solve are easiest to specify using multiple objectives in fact this is a borderline tautology an example which given our current sad situation is hard to escape is deciding what interventions a government should apply to limit the spread of a virus while maintaining economic 6 comments read more numerical errors perturbation analysis and machine learning get link facebook x pinterest email other apps september 16 2013 everyone hates numerical errors we love to think that computers are machines with infinite precision when i was a student i really hated error analysis it sounded like a subject that is set out to study an annoying side effect of our imperfect computers a boring detail that is miles away from anything that anyone would ever consider a nice part of mathematics i will not try to convince you today that the opposite is true however even in error analysis there are some nice ideas and lessons to be learned this post asks the question whether if you are doing machine learning you should care about numerical errors this issue should be well understood however i don t think that it is as well appreciated as it should be or that it received the attention it should in fact i doubt that the issue is discussed in any of the recent machine learning textbooks beyond the usual caveat beware the numerical errors scary in this blog i will illustrate the questi 1 comment read more student response systems a year later get link facebook x pinterest email other apps september 15 2013 i have promised to come back with my experience after the fall semester of 2013 though the end of that semester passed a long ago here are some thoughts overall my experience of socrative was very positive during the first week i polled the class to see how large a percentage of the class has some kind of wifi enabled device that they could use 90 out of the 95 students had some kind of device that they were willing to bring to the class so i decided to give socrative a try socrative helped me tremendously to stay on the top of what everyone in the class knows the way i used socrative was as follows after every block of new material usually 5 10 slides i inserted a few questions to verify whether the students got the new material this was all the extra work that i had to put in designing the class material due to socrative and i would have done something similar without socrative anyways so in fact i did not feel much of a difference here once t 3 comments read more student response systems get link facebook x pinterest email other apps september 10 2012 there are plenty student response systems srss out there which one to choose this brief document summarizes what i have found on the web before the fall of 2012 radio communication based systems the systems differ in a few characteristics the first is what type of devices the students can use classical iclicker like systems require the student buy a device which communicates with a receiver that the teacher should possess since our school uses iclickers let me focus on them the overall cost of first generation iclickers is 10 device assuming the student sells back the device to the bookstore which they can this is not a major cost but who likes to pay when you don t have to the first limitation of iclicker like systems is that they are bound to the smart classrooms and the computers there thus if you are like me and use your own computer for projection you will need to switch between screens to show the results of a poll this makes the use of icli post a comment read more more posts powered by blogger theme images by michael elkan csaba szepesvári visit profile archive mar 1 sept 2 sept 1 apr 1 may 1 apr 2 nov 2 oct 2 sept 1 apr 1 mar 3 aug 1 jul 1 may 2 apr 4 show more show less labels aggregation ai annoyance approximation theory artificial intelligence bayesian analysis bayesian models beamer style blogging clinical trials compression conditioning number constrained mdps curse of dimensionality djvu exploration frequentist approach galerkin s method hilbert matrix image processing jbig2 keynote latex learning theory least squares mac osx machine learning markovian decision processes mathematics matlab mdps mixing model selection models non parametric statistics occam s razor optimization tools optogenetics pdf perturbation analysis powerpoint presentation reinforcement learning representation learning reward hypothesis sample complexity science stability statistics supervised learning svn technical technology thunderbird tools x11 show more show less report abuse