hunch_net hunch_net-2006 hunch_net-2006-156 knowledge-graph by maker-knowledge-mining
Source: html
Introduction: I just visited Yahoo Research which has several fundamental learning problems near to (or beyond) the set of problems we know how to solve well. Here are 3 of them. Ranking This is the canonical problem of all search engines. It is made extra difficult for several reasons. There is relatively little “good” supervised learning data and a great deal of data with some signal (such as click through rates). The learning must occur in a partially adversarial environment. Many people very actively attempt to place themselves at the top of rankings. It is not even quite clear whether the problem should be posed as ‘ranking’ or as ‘regression’ which is then used to produce a ranking. Collaborative filtering Yahoo has a large number of recommendation systems for music, movies, etc… In these sorts of systems, users specify how they liked a set of things, and then the system can (hopefully) find some more examples of things they might like by reasoning across multiple
sentIndex sentText sentNum sentScore
1 I just visited Yahoo Research which has several fundamental learning problems near to (or beyond) the set of problems we know how to solve well. [sent-1, score-0.37]
2 Ranking This is the canonical problem of all search engines. [sent-3, score-0.361]
3 There is relatively little “good” supervised learning data and a great deal of data with some signal (such as click through rates). [sent-5, score-0.269]
4 The learning must occur in a partially adversarial environment. [sent-6, score-0.088]
5 Many people very actively attempt to place themselves at the top of rankings. [sent-7, score-0.206]
6 It is not even quite clear whether the problem should be posed as ‘ranking’ or as ‘regression’ which is then used to produce a ranking. [sent-8, score-0.098]
7 Exploration with Generalization The cash cow of search engines is displaying advertisements which are relevant to search along with search results. [sent-10, score-1.487]
8 Better targeting these advertisements makes money (a small improvement might be worth $millions) and improves the value of the search engine for the user. [sent-11, score-1.054]
9 It is natural to predict the set of advertisements which maximize the advertising payoff. [sent-12, score-0.823]
10 This natural idea is stymied by both the extreme multiplicity of advertisements under contract (think millions) and a lack of ability to measure hypotheticals like “What would have happened if we had displayed a different set of advertisements for this (query,user) pair instead? [sent-13, score-1.693]
11 ” This is a combined exploration and generalization problem. [sent-14, score-0.338]
12 Good solutions to any of these problems would be extremely useful (and not just at Yahoo). [sent-15, score-0.201]
13 Even further small improvements on the existing solutions may be very useful. [sent-16, score-0.181]
14 For those interested, Yahoo (as an organization) knows these are learning problems and is very actively interested in solving them. [sent-17, score-0.452]
15 Yahoo Research is committed to a relatively open method of solving these problems. [sent-18, score-0.247]
16 Dennis DeCoste is one contact point for machine learning research at Yahoo Research. [sent-19, score-0.175]
wordName wordTfidf (topN-words)
[('advertisements', 0.461), ('yahoo', 0.368), ('search', 0.282), ('millions', 0.184), ('generalization', 0.142), ('ranking', 0.136), ('actively', 0.129), ('exploration', 0.123), ('set', 0.111), ('solutions', 0.109), ('stymied', 0.105), ('displayed', 0.105), ('pair', 0.105), ('posed', 0.098), ('multiplicity', 0.098), ('contract', 0.098), ('movies', 0.098), ('research', 0.093), ('problems', 0.092), ('signal', 0.092), ('engines', 0.092), ('click', 0.092), ('occur', 0.088), ('maximize', 0.088), ('cash', 0.088), ('systems', 0.085), ('relatively', 0.085), ('reasoning', 0.084), ('engine', 0.084), ('targeting', 0.084), ('advertising', 0.084), ('collaborative', 0.084), ('committed', 0.084), ('contact', 0.082), ('natural', 0.079), ('canonical', 0.079), ('knows', 0.079), ('solving', 0.078), ('top', 0.077), ('organization', 0.075), ('recommendation', 0.075), ('visited', 0.075), ('interested', 0.074), ('combined', 0.073), ('filtering', 0.073), ('liked', 0.073), ('small', 0.072), ('users', 0.071), ('improves', 0.071), ('happened', 0.07)]
simIndex simValue blogId blogTitle
same-blog 1 1.0 156 hunch net-2006-02-11-Yahoo’s Learning Problems.
Introduction: I just visited Yahoo Research which has several fundamental learning problems near to (or beyond) the set of problems we know how to solve well. Here are 3 of them. Ranking This is the canonical problem of all search engines. It is made extra difficult for several reasons. There is relatively little “good” supervised learning data and a great deal of data with some signal (such as click through rates). The learning must occur in a partially adversarial environment. Many people very actively attempt to place themselves at the top of rankings. It is not even quite clear whether the problem should be posed as ‘ranking’ or as ‘regression’ which is then used to produce a ranking. Collaborative filtering Yahoo has a large number of recommendation systems for music, movies, etc… In these sorts of systems, users specify how they liked a set of things, and then the system can (hopefully) find some more examples of things they might like by reasoning across multiple
2 0.30142906 175 hunch net-2006-04-30-John Langford –> Yahoo Research, NY
Introduction: I will join Yahoo Research (in New York) after my contract ends at TTI-Chicago . The deciding reasons are: Yahoo is running into many hard learning problems. This is precisely the situation where basic research might hope to have the greatest impact. Yahoo Research understands research including publishing, conferences, etc… Yahoo Research is growing, so there is a chance I can help it grow well. Yahoo understands the internet, including (but not at all limited to) experimenting with research blogs. In the end, Yahoo Research seems like the place where I might have a chance to make the greatest difference. Yahoo (as a company) has made a strong bet on Yahoo Research. We-the-researchers all hope that bet will pay off, and this seems plausible. I’ll certainly have fun trying.
3 0.24615458 178 hunch net-2006-05-08-Big machine learning
Introduction: According to the New York Times , Yahoo is releasing Project Panama shortly . Project Panama is about better predicting which advertisements are relevant to a search, implying a higher click through rate, implying larger income for Yahoo . There are two things that seem interesting here: A significant portion of that improved accuracy is almost certainly machine learning at work. The quantitative effect is huge—the estimate in the article is $600*10 6 . Google already has such improvements and Microsoft Search is surely working on them, which suggest this is (perhaps) a $10 9 per year machine learning problem. The exact methodology under use is unlikely to be publicly discussed in the near future because of the competitive enivironment. Hopefully we’ll have some public “war stories” at some point in the future when this information becomes less sensitive. For now, it’s reassuring to simply note that machine learning is having a big impact.
4 0.17477106 423 hunch net-2011-02-02-User preferences for search engines
Introduction: I want to comment on the “Bing copies Google” discussion here , here , and here , because there are data-related issues which the general public may not understand, and some of the framing seems substantially misleading to me. As a not-distant-outsider, let me mention the sources of bias I may have. I work at Yahoo! , which has started using Bing . This might predispose me towards Bing, but on the other hand I’m still at Yahoo!, and have been using Linux exclusively as an OS for many years, including even a couple minor kernel patches. And, on the gripping hand , I’ve spent quite a bit of time thinking about the basic principles of incorporating user feedback in machine learning . Also note, this post is not related to official Yahoo! policy, it’s just my personal view. The issue Google engineers inserted synthetic responses to synthetic queries on google.com, then executed the synthetic searches on google.com using Internet Explorer with the Bing toolbar and later
5 0.17184018 120 hunch net-2005-10-10-Predictive Search is Coming
Introduction: “Search” is the other branch of AI research which has been succesful. Concrete examples include Deep Blue which beat the world chess champion and Chinook the champion checkers program. A set of core search techniques exist including A * , alpha-beta pruning, and others that can be applied to any of many different search problems. Given this, it may be surprising to learn that there has been relatively little succesful work on combining prediction and search. Given also that humans typically solve search problems using a number of predictive heuristics to narrow in on a solution, we might be surprised again. However, the big successful search-based systems have typically not used “smart” search algorithms. Insteady they have optimized for very fast search. This is not for lack of trying… many people have tried to synthesize search and prediction to various degrees of success. For example, Knightcap achieves good-but-not-stellar chess playing performance, and TD-gammon
6 0.16628957 464 hunch net-2012-05-03-Microsoft Research, New York City
7 0.16348863 200 hunch net-2006-08-03-AOL’s data drop
8 0.12948102 389 hunch net-2010-02-26-Yahoo! ML events
9 0.1189247 132 hunch net-2005-11-26-The Design of an Optimal Research Environment
10 0.11846599 269 hunch net-2007-10-24-Contextual Bandits
11 0.11597265 345 hunch net-2009-03-08-Prediction Science
12 0.11319853 425 hunch net-2011-02-25-Yahoo! Machine Learning grant due March 11
13 0.1120394 444 hunch net-2011-09-07-KDD and MUCMD 2011
14 0.1062197 364 hunch net-2009-07-11-Interesting papers at KDD
15 0.10333464 22 hunch net-2005-02-18-What it means to do research.
16 0.10255069 400 hunch net-2010-06-13-The Good News on Exploration and Learning
17 0.097792082 435 hunch net-2011-05-16-Research Directions for Machine Learning and Algorithms
18 0.094224446 260 hunch net-2007-08-25-The Privacy Problem
19 0.09259782 359 hunch net-2009-06-03-Functionally defined Nonlinear Dynamic Models
20 0.085778125 378 hunch net-2009-11-15-The Other Online Learning
topicId topicWeight
[(0, 0.205), (1, 0.025), (2, -0.113), (3, 0.094), (4, -0.071), (5, -0.045), (6, 0.002), (7, 0.067), (8, -0.079), (9, -0.041), (10, 0.002), (11, 0.145), (12, -0.061), (13, 0.113), (14, -0.141), (15, 0.146), (16, -0.087), (17, -0.064), (18, 0.037), (19, -0.153), (20, 0.112), (21, 0.07), (22, 0.107), (23, -0.024), (24, 0.019), (25, 0.2), (26, 0.124), (27, -0.016), (28, -0.066), (29, 0.031), (30, -0.071), (31, 0.017), (32, -0.099), (33, -0.011), (34, 0.058), (35, 0.097), (36, 0.033), (37, -0.012), (38, 0.026), (39, 0.03), (40, 0.041), (41, -0.051), (42, -0.059), (43, -0.047), (44, 0.097), (45, -0.019), (46, 0.059), (47, 0.136), (48, 0.022), (49, 0.013)]
simIndex simValue blogId blogTitle
same-blog 1 0.95299762 156 hunch net-2006-02-11-Yahoo’s Learning Problems.
Introduction: I just visited Yahoo Research which has several fundamental learning problems near to (or beyond) the set of problems we know how to solve well. Here are 3 of them. Ranking This is the canonical problem of all search engines. It is made extra difficult for several reasons. There is relatively little “good” supervised learning data and a great deal of data with some signal (such as click through rates). The learning must occur in a partially adversarial environment. Many people very actively attempt to place themselves at the top of rankings. It is not even quite clear whether the problem should be posed as ‘ranking’ or as ‘regression’ which is then used to produce a ranking. Collaborative filtering Yahoo has a large number of recommendation systems for music, movies, etc… In these sorts of systems, users specify how they liked a set of things, and then the system can (hopefully) find some more examples of things they might like by reasoning across multiple
2 0.81494606 178 hunch net-2006-05-08-Big machine learning
Introduction: According to the New York Times , Yahoo is releasing Project Panama shortly . Project Panama is about better predicting which advertisements are relevant to a search, implying a higher click through rate, implying larger income for Yahoo . There are two things that seem interesting here: A significant portion of that improved accuracy is almost certainly machine learning at work. The quantitative effect is huge—the estimate in the article is $600*10 6 . Google already has such improvements and Microsoft Search is surely working on them, which suggest this is (perhaps) a $10 9 per year machine learning problem. The exact methodology under use is unlikely to be publicly discussed in the near future because of the competitive enivironment. Hopefully we’ll have some public “war stories” at some point in the future when this information becomes less sensitive. For now, it’s reassuring to simply note that machine learning is having a big impact.
3 0.75532019 423 hunch net-2011-02-02-User preferences for search engines
Introduction: I want to comment on the “Bing copies Google” discussion here , here , and here , because there are data-related issues which the general public may not understand, and some of the framing seems substantially misleading to me. As a not-distant-outsider, let me mention the sources of bias I may have. I work at Yahoo! , which has started using Bing . This might predispose me towards Bing, but on the other hand I’m still at Yahoo!, and have been using Linux exclusively as an OS for many years, including even a couple minor kernel patches. And, on the gripping hand , I’ve spent quite a bit of time thinking about the basic principles of incorporating user feedback in machine learning . Also note, this post is not related to official Yahoo! policy, it’s just my personal view. The issue Google engineers inserted synthetic responses to synthetic queries on google.com, then executed the synthetic searches on google.com using Internet Explorer with the Bing toolbar and later
4 0.7235328 200 hunch net-2006-08-03-AOL’s data drop
Introduction: AOL has released several large search engine related datasets. This looks like a pretty impressive data release, and it is a big opportunity for people everywhere to worry about search engine related learning problems, if they want.
5 0.72101867 175 hunch net-2006-04-30-John Langford –> Yahoo Research, NY
Introduction: I will join Yahoo Research (in New York) after my contract ends at TTI-Chicago . The deciding reasons are: Yahoo is running into many hard learning problems. This is precisely the situation where basic research might hope to have the greatest impact. Yahoo Research understands research including publishing, conferences, etc… Yahoo Research is growing, so there is a chance I can help it grow well. Yahoo understands the internet, including (but not at all limited to) experimenting with research blogs. In the end, Yahoo Research seems like the place where I might have a chance to make the greatest difference. Yahoo (as a company) has made a strong bet on Yahoo Research. We-the-researchers all hope that bet will pay off, and this seems plausible. I’ll certainly have fun trying.
6 0.63760829 464 hunch net-2012-05-03-Microsoft Research, New York City
7 0.61745155 120 hunch net-2005-10-10-Predictive Search is Coming
8 0.5110395 389 hunch net-2010-02-26-Yahoo! ML events
9 0.50336647 339 hunch net-2009-01-27-Key Scientific Challenges
10 0.44863236 424 hunch net-2011-02-17-What does Watson mean?
11 0.41985422 345 hunch net-2009-03-08-Prediction Science
12 0.41620618 117 hunch net-2005-10-03-Not ICML
13 0.41541085 31 hunch net-2005-02-26-Problem: Reductions and Relative Ranking Metrics
14 0.40661141 425 hunch net-2011-02-25-Yahoo! Machine Learning grant due March 11
15 0.40631452 400 hunch net-2010-06-13-The Good News on Exploration and Learning
16 0.40556112 269 hunch net-2007-10-24-Contextual Bandits
17 0.39956829 132 hunch net-2005-11-26-The Design of an Optimal Research Environment
18 0.39919004 260 hunch net-2007-08-25-The Privacy Problem
19 0.39429215 457 hunch net-2012-02-29-Key Scientific Challenges and the Franklin Symposium
20 0.38931203 18 hunch net-2005-02-12-ROC vs. Accuracy vs. AROC
topicId topicWeight
[(10, 0.027), (27, 0.202), (38, 0.068), (53, 0.042), (55, 0.082), (68, 0.3), (79, 0.03), (94, 0.127), (95, 0.029)]
simIndex simValue blogId blogTitle
1 0.92572832 48 hunch net-2005-03-29-Academic Mechanism Design
Introduction: From game theory, there is a notion of “mechanism design”: setting up the structure of the world so that participants have some incentive to do sane things (rather than obviously counterproductive things). Application of this principle to academic research may be fruitful. What is misdesigned about academic research? The JMLG guides give many hints. The common nature of bad reviewing also suggests the system isn’t working optimally. There are many ways to experimentally “cheat” in machine learning . Funding Prisoner’s Delimma. Good researchers often write grant proposals for funding rather than doing research. Since the pool of grant money is finite, this means that grant proposals are often rejected, implying that more must be written. This is essentially a “prisoner’s delimma”: anyone not writing grant proposals loses, but the entire process of doing research is slowed by distraction. If everyone wrote 1/2 as many grant proposals, roughly the same distribution
2 0.88920462 428 hunch net-2011-03-27-Vowpal Wabbit, v5.1
Introduction: I just created version 5.1 of vowpal wabbit . This almost entirely a bugfix release, so it’s an easy upgrade from v5.0. In addition: There is now a mailing list , which I and several other developers are subscribed to. The main website has shifted to the wiki on github. This means that anyone with a github account can now edit it. I’m planning to give a tutorial tomorrow on it at eHarmony / the LA machine learning meetup at 10am. Drop by if you’re interested. The status of VW amongst other open source projects has changed. When VW first came out, it was relatively unique amongst existing projects in terms of features. At this point, many other projects have started to appreciate the value of the design choices here. This includes: Mahout , which now has an SGD implementation. Shogun , where Soeren is keen on incorporating features . LibLinear , where they won the KDD best paper award for out-of-core learning . This is expected—any open sourc
3 0.88536799 110 hunch net-2005-09-10-“Failure” is an option
Introduction: This is about the hard choices that graduate students must make. The cultural definition of success in academic research is to: Produce good research which many other people appreciate. Produce many students who go on to do the same. There are fundamental reasons why this is success in the local culture. Good research appreciated by others means access to jobs. Many students succesful in the same way implies that there are a number of people who think in a similar way and appreciate your work. In order to graduate, a phd student must live in an academic culture for a period of several years. It is common to adopt the culture’s definition of success during this time. It’s also common for many phd students discover they are not suited to an academic research lifestyle. This collision of values and abilities naturally results in depression. The most fundamental advice when this happens is: change something. Pick a new advisor. Pick a new research topic. Or leave th
same-blog 4 0.87611037 156 hunch net-2006-02-11-Yahoo’s Learning Problems.
Introduction: I just visited Yahoo Research which has several fundamental learning problems near to (or beyond) the set of problems we know how to solve well. Here are 3 of them. Ranking This is the canonical problem of all search engines. It is made extra difficult for several reasons. There is relatively little “good” supervised learning data and a great deal of data with some signal (such as click through rates). The learning must occur in a partially adversarial environment. Many people very actively attempt to place themselves at the top of rankings. It is not even quite clear whether the problem should be posed as ‘ranking’ or as ‘regression’ which is then used to produce a ranking. Collaborative filtering Yahoo has a large number of recommendation systems for music, movies, etc… In these sorts of systems, users specify how they liked a set of things, and then the system can (hopefully) find some more examples of things they might like by reasoning across multiple
5 0.78026098 345 hunch net-2009-03-08-Prediction Science
Introduction: One view of machine learning is that it’s about how to program computers to predict well. This suggests a broader research program centered around the more pervasive goal of simply predicting well. There are many distinct strands of this broader research program which are only partially unified. Here are the ones that I know of: Learning Theory . Learning theory focuses on several topics related to the dynamics and process of prediction. Convergence bounds like the VC bound give an intellectual foundation to many learning algorithms. Online learning algorithms like Weighted Majority provide an alternate purely game theoretic foundation for learning. Boosting algorithms yield algorithms for purifying prediction abiliity. Reduction algorithms provide means for changing esoteric problems into well known ones. Machine Learning . A great deal of experience has accumulated in practical algorithm design from a mixture of paradigms, including bayesian, biological, opt
6 0.72051138 65 hunch net-2005-05-02-Reviewing techniques for conferences
7 0.65439647 132 hunch net-2005-11-26-The Design of an Optimal Research Environment
8 0.65084016 225 hunch net-2007-01-02-Retrospective
9 0.64922565 286 hunch net-2008-01-25-Turing’s Club for Machine Learning
10 0.64921367 95 hunch net-2005-07-14-What Learning Theory might do
11 0.6479184 207 hunch net-2006-09-12-Incentive Compatible Reviewing
12 0.64761615 411 hunch net-2010-09-21-Regretting the dead
13 0.64523643 366 hunch net-2009-08-03-Carbon in Computer Science Research
14 0.644952 358 hunch net-2009-06-01-Multitask Poisoning
15 0.64150804 221 hunch net-2006-12-04-Structural Problems in NIPS Decision Making
16 0.64149064 449 hunch net-2011-11-26-Giving Thanks
17 0.64113116 204 hunch net-2006-08-28-Learning Theory standards for NIPS 2006
18 0.63979518 382 hunch net-2009-12-09-Future Publication Models @ NIPS
19 0.6381864 359 hunch net-2009-06-03-Functionally defined Nonlinear Dynamic Models
20 0.63801658 423 hunch net-2011-02-02-User preferences for search engines