WEBVTT 1 00:00:00.000 --> 00:00:01.589 Matthew Bellis: get a group photo. 2 00:00:03.330 --> 00:00:12.599 Matthew Bellis: That group photo is going to be a screenshot of probably multiple pages of our of our zoom session. 3 00:00:13.799 --> 00:00:23.400 Matthew Bellis: So that'll happen again, just before the second break we'll ask everybody if you're comfortable to turn on your cameras at least just for 60 seconds or so. 4 00:00:24.090 --> 00:00:39.060 Matthew Bellis: So that gives you time if you want to if you need to change what you're wearing i've just got the basic black shirt or anybody wants to set a green screen behind them, but again we'll do that just before the second break okay. 5 00:00:40.140 --> 00:00:48.900 Matthew Bellis: Alright, so i'm going to go ahead and jump right in, and I will start with the main page so First let me share my screen. 6 00:00:50.040 --> 00:00:52.710 Matthew Bellis: Give me one moment share screen. 7 00:00:55.980 --> 00:00:56.850 Matthew Bellis: See, I think. 8 00:00:57.870 --> 00:01:01.860 Matthew Bellis: No, I keep sharing pamphlet accidentally all right. 9 00:01:03.210 --> 00:01:05.520 Matthew Bellis: Back to figure out which one to share. 10 00:01:07.800 --> 00:01:08.520 Matthew Bellis: See. 11 00:01:15.480 --> 00:01:16.200 This is it. 12 00:01:17.250 --> 00:01:29.490 Matthew Bellis: There we go okay so just to walk through where to get to what i'm about to show you if you start on the main workshop website, you can go to timetable. 13 00:01:33.600 --> 00:01:52.170 Matthew Bellis: You click on time tape you'll see a breakdown, not just of yesterday Monday July 19 and if you scroll further down you get to Tuesday will be starting off here with data sets and access I have set my time to be local so your time may show up differently. 14 00:01:54.210 --> 00:02:00.570 Matthew Bellis: And then, just to point out here's the first break and then the second break just before that is we're will show the photos. 15 00:02:02.310 --> 00:02:03.510 Matthew Bellis: So if you. 16 00:02:04.950 --> 00:02:14.610 Matthew Bellis: start here data sets online three solutions and questions, there is a new link to Google slides solutions to offline. 17 00:02:15.900 --> 00:02:16.710 challenge. 18 00:02:24.240 --> 00:02:33.810 Matthew Bellis: And I will go through a little bit quicker the hands on activity from yesterday i've tried to provide links to everything so people can follow up with this. 19 00:02:35.640 --> 00:02:48.090 Matthew Bellis: And then we'll go through the the offline challenge from yesterday and maybe there'll be some discussion about both the physics and accessing these data, let me go into present mode. 20 00:02:50.340 --> 00:02:50.970 Matthew Bellis: and 21 00:02:52.140 --> 00:03:04.470 Matthew Bellis: Let me turn on my laser pointer could do Okay, and if if anybody could give like a thumbs up or anything if they see my screen, I just want to make sure i'm actually sharing the right screen. 22 00:03:07.080 --> 00:03:15.660 Matthew Bellis: Thank you very much, all right perfect awesome Thank you manual alright so let's go through these and there was a hands on activity yesterday. 23 00:03:16.650 --> 00:03:29.160 Matthew Bellis: It was basically reviewing the concepts trying to encourage people to go back through and read a little bit more closely in the event that I may have gone too fast i'll try to move this out of the way. 24 00:03:30.870 --> 00:03:36.960 Matthew Bellis: If I went too fast, just to remind people how to get to the starting page how you select the cms data. 25 00:03:38.190 --> 00:03:48.960 Matthew Bellis: You know, reminding people, some of the ideas about what is the energy, the difference between collision and simulated data selecting a God and then add simulated data. 26 00:03:50.220 --> 00:03:57.000 Matthew Bellis: Just reminding people how to get to maybe some of the specific processes and how to understand the information. 27 00:03:58.200 --> 00:04:09.540 Matthew Bellis: And as we went through that we found out that some of these data sets are actually not accessible, we have records for them so that if people wanted to they could request to access to them. 28 00:04:09.990 --> 00:04:21.750 Matthew Bellis: But it means there may be some limited information if you want to know the details of how they were produced, but, in all events, you can always contact the open data organizers. 29 00:04:22.980 --> 00:04:24.120 Matthew Bellis: Through the web pages. 30 00:04:26.700 --> 00:04:33.510 Matthew Bellis: So again, just to kind of walk you through what these slides are for the hands on activity. 31 00:04:34.020 --> 00:04:43.380 Matthew Bellis: Can you get to the starting page, can you select cms data I provided links just remind people where the portal is and then, if you select cms. 32 00:04:43.590 --> 00:04:54.420 Matthew Bellis: You come to a page that looks something like this, this link is just accessing that you can see that experimental cms has been selected, but of course you don't need to type out the full. 33 00:04:56.310 --> 00:05:00.210 Matthew Bellis: link, you can just select it by going to cms here. 34 00:05:03.030 --> 00:05:24.900 Matthew Bellis: The bulk of the deer cms data was released for the 2010 2011 and 2012 run periods that was run one the collision energies, for these are seven or eight TV and as Connie mentioned, I think, in the opening session there will be data coming for run to the 13 TV collisions. 35 00:05:27.240 --> 00:05:40.020 Matthew Bellis: difference between collision and simulated data collision data is I use this jargon, even though it's probably not the right one to us it's the real data that comes off of the real cms detector not the simulated detector. 36 00:05:40.500 --> 00:05:53.700 Matthew Bellis: simulated detectors Monte Carlo data we've actually simulated not just the underlying physics process of defense of scale, but what these particles are doing as they move through a simulation of the cms detector. 37 00:05:54.360 --> 00:06:07.890 Matthew Bellis: And at some point, both of these data sets look the same because they've gone through the same process, and so you can use your same analysis software to look for electrons, for example in both collision data or simulated data. 38 00:06:13.530 --> 00:06:26.490 Matthew Bellis: to remind people what the ao D format is that's analysis object data emphasis on the analysis, this was supposed to be a level of processing that most analysts would be able to use. 39 00:06:28.020 --> 00:06:47.520 Matthew Bellis: For their studies, it has a walk of information and even within the collaboration we tend to use smaller slimmed down versions of a God, but for the open data since people may be looking at, for many, many different processes many different types of studies. 40 00:06:48.990 --> 00:07:07.140 Matthew Bellis: It was decided to release the kind of more complete version of this process data if you go through this we discussed that this first field in between the slashes references the trigger so mule EG new had single photon. 41 00:07:08.310 --> 00:07:14.970 Matthew Bellis: and so on, and if you click on these links, you can find out what they're doing if you go into any one of these. 42 00:07:15.900 --> 00:07:26.250 Matthew Bellis: You get a bit more information, so if you were to click on this one, this is the meal EG primary data set, how are these data selected all of these records have a field. 43 00:07:26.850 --> 00:07:38.370 Matthew Bellis: Events were selected because of the presence of two different types of high energy particles, such as a meal on an electron or photon, and the reason why it's an electron or photon is because you're looking. 44 00:07:40.380 --> 00:07:52.980 Matthew Bellis: you're looking for energy deposited in a detector electromagnetic Keller emitter that's a first order when you're actually just triggering on it, you don't know whether or not it was an electron or photon you know that it was something that. 45 00:07:54.150 --> 00:08:02.610 Matthew Bellis: tends to interact electromagnetically and then at later processing, you can decide whether or not it wasn't electron or photon. 46 00:08:05.730 --> 00:08:10.320 Matthew Bellis: I asked you to select the heavy gauge bosoms buying the Z prime sample. 47 00:08:11.130 --> 00:08:17.280 Matthew Bellis: So if you click the on demand data sets if you scroll further down you find this filter by category. 48 00:08:17.580 --> 00:08:30.180 Matthew Bellis: And you find these very specific business processes various exotica standard model of physics hates physics, both the standard model pigs and beyond the standard model, and you can select heavy gauge pose ons. 49 00:08:31.110 --> 00:08:40.950 Matthew Bellis: and find that there's a the prime samples, we went through this and found out that maybe not all of the information is a is available. 50 00:08:41.940 --> 00:09:00.180 Matthew Bellis: You can learn more about the naming schemes of these all of the records have a link to about cms simulated data sets where you can learn how we tried to label these this is in some sense some of the provenance how it was processed hit the a six till for any tower decays. 51 00:09:01.380 --> 00:09:05.580 Matthew Bellis: As well as information about like the mouse range in this example, right here. 52 00:09:07.290 --> 00:09:14.250 Matthew Bellis: And you can also find out how much how many events ran that sample and how much disk space it occupies. 53 00:09:14.850 --> 00:09:31.080 Matthew Bellis: So in this particular example there are 950 6164 events, it occupies 361 gigabytes, and this should give you some sense of the amount of data that's produced this is one very specific physics process just the simulated data. 54 00:09:31.560 --> 00:09:39.030 Matthew Bellis: And it's already 360 gigabytes so it's a significant amount of data that you have access to. 55 00:09:41.760 --> 00:09:55.080 Matthew Bellis: i'm going to see something up here I can't see the chat as much um so i'm going to ask the other facilitators to just if you feel that they're if there's a question. 56 00:09:55.620 --> 00:10:09.120 Matthew Bellis: That I have missed if one of you can just like chime in and interrupt me and we can address it, but I just saw that there is chats but for some reason i'm not able to access the chat right now. 57 00:10:10.530 --> 00:10:16.380 Matthew Bellis: So again, if any of the other facilitators feel or something I should stop just interrupt me i'm good with that. 58 00:10:20.250 --> 00:10:24.870 Matthew Bellis: And there is a perhaps more information and some of the data sets for the La G files. 59 00:10:26.550 --> 00:10:35.520 Matthew Bellis: But you may not be able to access all of the loose files that's really the tree level almost five Min diagram points to the process. 60 00:10:37.170 --> 00:10:37.560 Matthew Bellis: Okay. 61 00:10:39.240 --> 00:10:43.650 Matthew Bellis: i'm going to pause there for a second i'm going to stop sharing so that I can see the chat. 62 00:10:45.600 --> 00:10:49.530 Matthew Bellis: To see if there are any questions at this point. 63 00:10:54.930 --> 00:11:03.840 Matthew Bellis: There was a question about the on demand data sets and, yes, I believe that those are always the the simulated data as Julie said as well and that's my understanding. 64 00:11:07.560 --> 00:11:09.120 Matthew Bellis: Any other questions about that. 65 00:11:13.140 --> 00:11:15.960 Matthew Bellis: All right back into the fray. 66 00:11:17.190 --> 00:11:19.140 Matthew Bellis: let's go back let's see if there we go. 67 00:11:22.200 --> 00:11:27.450 Matthew Bellis: there's a question about what does the you mean in the prime T to TT you. 68 00:11:28.080 --> 00:11:37.740 Matthew Bellis: Do that's a great great question, and when I was putting this exercise together, I spent a fair amount of time trying to figure out, what do you mean, and I could not do it. 69 00:11:38.310 --> 00:11:55.620 Matthew Bellis: that's when I wound up trying to find the files or even dump the generated data and that's when I realized like Oh, this is an on demand data set actually can't access that real low level information, so I don't know what the EU is. 70 00:11:57.360 --> 00:12:04.500 Matthew Bellis: I would defer to any of my colleagues who know what the you means in Z prime T to TT you. 71 00:12:08.010 --> 00:12:08.700 Matthew Bellis: Anybody know. 72 00:12:10.560 --> 00:12:19.800 Matthew Bellis: Sorry Jerry you read the the edge of my knowledge, and someone said, the data sets are gauge data is there any online. 73 00:12:20.910 --> 00:12:29.280 Matthew Bellis: cloud based space where we can work on data sets without downloading this huge data and my laptop yes, through the virtual. 74 00:12:31.440 --> 00:12:36.990 Matthew Bellis: Environment you could skim subsets of the data so that you only have to download smaller subsets. 75 00:12:37.800 --> 00:12:50.400 Matthew Bellis: But on Thursday, you will be learning about how to do analysis in the cloud and so that is Wednesday night Wednesday night for me, you will be preparing. 76 00:12:51.060 --> 00:13:00.930 Matthew Bellis: For a tutorial on on Thursday where you're going to do this on Google cloud and it's going to feel like you're part of the future it's kind of amazing. 77 00:13:01.980 --> 00:13:03.030 Matthew Bellis: So good question here. 78 00:13:04.890 --> 00:13:08.340 Matthew Bellis: And let me again apologize for anybody whose names I miss pronounce. 79 00:13:09.540 --> 00:13:12.930 Matthew Bellis: feel free to correct me if if I really mangle it. 80 00:13:14.640 --> 00:13:15.720 Matthew Bellis: You have a questions. 81 00:13:17.850 --> 00:13:22.500 Matthew Bellis: Okay, let me jump back into this solutions. 82 00:13:24.150 --> 00:13:45.150 Matthew Bellis: OK, so the offline challenge a positive this situation where you have come up with a new theory is ultimate on you want to see if you can discover it in the data and then you're going to have this like big colloquium seminar at at CERN showing the results of your discovery. 83 00:13:46.410 --> 00:13:50.370 Matthew Bellis: This particle the ultimate Tom said, we can move this out of the way. 84 00:13:52.350 --> 00:14:05.100 Matthew Bellis: Will decay to to same sign charged leptons so those electrons neutrons are towels and to jets have any flavors so that means they could be be jets, but they can also be like court jets they could be. 85 00:14:06.060 --> 00:14:16.410 Matthew Bellis: They could be charm strange what have you the predicted massive dust particles and the 300 to 500 G the range you plan on looking for it in the cms open data. 86 00:14:16.920 --> 00:14:25.800 Matthew Bellis: Only considered the 2012 run period and the first thing you want to think about are what triggers are sensitive to this process, how many events might that give you. 87 00:14:26.640 --> 00:14:42.960 Matthew Bellis: What standard model processes might contribute to the background so we're going to talk about physics, for a bit, because it really does map on to how you use the open data, and then I give you a very specific sample and to try to consider if that's a significant background. 88 00:14:44.160 --> 00:14:49.920 Matthew Bellis: And the first question, you want to think about is what triggers might be most sensitive to this process. 89 00:14:50.760 --> 00:15:05.460 Matthew Bellis: We are not going to be able to teach you everything in this workshop we're not going to be able to teach you everything about the physics, this is why you know particle physics PhDs take four or five years, sometimes more. 90 00:15:07.080 --> 00:15:21.750 Matthew Bellis: But if you look at this process, the way that I would have written down is your ultimate on particle decays to to same sign leptons so they could both be plus they could both be minus let me get my little laser pointer back on. 91 00:15:22.500 --> 00:15:29.340 Matthew Bellis: And then to jet so the way I would tend to write this down is left on leptons jet jet. 92 00:15:30.600 --> 00:15:41.130 Matthew Bellis: And so, our triggers you probably want to trigger that's looking forward to leptons there are some triggers where you might have like a large amounts of like had chronic deposition. 93 00:15:42.060 --> 00:15:47.670 Matthew Bellis: But probably the easiest thing to pick out is two or more charts leptons. 94 00:15:48.120 --> 00:15:57.210 Matthew Bellis: And there's a few ways, you can approach this there is, if you poke around the hope was that people would go through some of these triggers and read a little bit more about them. 95 00:15:57.690 --> 00:16:04.860 Matthew Bellis: There is the new EG they are selected because of the presence of two different types of particles, such as a new on and an electron. 96 00:16:05.550 --> 00:16:17.520 Matthew Bellis: The double new parked at least two high energy nuance the double electronic these two high energy electrons and then there's the town park as well, we have the presence of more than one isolated TAO. 97 00:16:17.910 --> 00:16:22.830 Matthew Bellis: With maybe an energetic chat or at least one isolated town high missing transverse energy. 98 00:16:23.940 --> 00:16:30.180 Matthew Bellis: Now these these triggers do not guarantee that you will also have to jets. 99 00:16:30.960 --> 00:16:46.350 Matthew Bellis: But it is not none of these triggers are eliminating any data that have jets so what ideally you get is events that have at least two leptons there's nothing in this about whether or not they are saying sign or not. 100 00:16:48.270 --> 00:17:00.780 Matthew Bellis: And then within them you EG data set, you may look for events, then that are plus plus or minus minus for the laptop plus two or more checks and then you could start looking at things like the. 101 00:17:01.860 --> 00:17:05.700 Matthew Bellis: You can start looking at things like the invariant mass or anything like that. 102 00:17:06.630 --> 00:17:23.100 Matthew Bellis: But this is the this is how you start thinking about about your your triggers again this trigger is not going to get exactly what you want, but you don't want it to eliminate necessarily anything you want these to all contain at least your data. 103 00:17:26.790 --> 00:17:39.000 Matthew Bellis: So um if you wanted to know how many records, there are you can click on them and what you find is that for 2012 there were to run periods that have been released, there is a be. 104 00:17:39.720 --> 00:18:02.220 Matthew Bellis: run period and a see reading period and it looks like I should put them in more consistent order so again for 2012 for each of these triggers or two sub run periods, there is this see January 22 2013 which is probably when it was processed and this be 22nd January 2013 or version one. 105 00:18:04.680 --> 00:18:13.020 Matthew Bellis: There are 14 million events and one of them 20 million events for the other just for the mute EG sample. 106 00:18:14.790 --> 00:18:25.530 Matthew Bellis: That occupies four and six terabytes of data, respectively, and which goes to the question that was raised earlier, you do not want to have to download this to your laptop. 107 00:18:26.220 --> 00:18:34.920 Matthew Bellis: Instead, what you want to do is you want to do your processing either small amounts of it that you download to your laptop so some very restrictive analysis. 108 00:18:35.310 --> 00:18:42.630 Matthew Bellis: Using the the physical objects that Julie, will be talking about later, or you want to set up a cloud computing environment that you'll learn on Thursday. 109 00:18:46.110 --> 00:19:00.960 Matthew Bellis: So you see that there is right so here's about 34 million events here is about 66 million events, this is again on the order of 56 and so on, this is 1600 over 100 million events. 110 00:19:02.340 --> 00:19:07.380 Matthew Bellis: However, you have to be careful that you do not double count your data. 111 00:19:09.030 --> 00:19:18.750 Matthew Bellis: And again, we cannot teach you everything in this workshop, but I will point out one thing to be aware of some events may pass two or more triggers. 112 00:19:20.310 --> 00:19:39.270 Matthew Bellis: So events that have to mute ons or two electrons those generally do not appear those generally do not overlap, although you could have some events with to me ones and two electrons but, for example, this mew EG sample here, this is looking for at least one meal on and at least one electron. 113 00:19:40.350 --> 00:19:51.450 Matthew Bellis: If something pass a trigger whether to nuance there could also have been an electron in this sample So these are not exclusive triggers. 114 00:19:53.550 --> 00:20:03.210 Matthew Bellis: Generally, what you will find analysts do is they pick one maybe two triggers that they will use for this analysis and then use that trigger for both their data set. 115 00:20:04.470 --> 00:20:12.930 Matthew Bellis: And the Monte Carlo and if you're using two triggers again, you want to be careful that you're not doing something that might accidentally over count. 116 00:20:14.850 --> 00:20:26.940 Matthew Bellis: And so you will see here this theme throughout the workshop it's really up to you to determine what data samples to use how to optimize your data, what are the best triggers. 117 00:20:27.840 --> 00:20:51.030 Matthew Bellis: That is the the work of the analysts, there is no one size fits all approach, there is no one and only one way to optimize your data that's where that's where the fun comes in, is trying to solve all these issues, but I think it's important to note that these triggers are not always mutually. 118 00:20:52.320 --> 00:20:53.070 Matthew Bellis: exclusive. 119 00:20:54.690 --> 00:20:59.220 Matthew Bellis: And that's actually kind of an important point i'm going to stop sharing for just a second. 120 00:21:01.260 --> 00:21:07.290 Matthew Bellis: Is there any questions about that point, or what I mean when I say that these are not exclusive. 121 00:21:09.330 --> 00:21:10.290 Matthew Bellis: processes. 122 00:21:12.750 --> 00:21:18.210 Matthew Bellis: Oh, there was actually a discussion about that and it looks like I said, is there a thank you Cathy for answering that. 123 00:21:28.380 --> 00:21:28.800 Matthew Bellis: Right. 124 00:21:29.820 --> 00:21:31.440 Matthew Bellis: back into this. 125 00:21:40.350 --> 00:21:49.830 Matthew Bellis: Okay, so what background might process I I know some of you are more familiar with the physic processes than others, so I thought it would be. 126 00:21:50.280 --> 00:21:53.250 Matthew Bellis: useful to kind of review what the physics is. 127 00:21:53.610 --> 00:22:06.720 Matthew Bellis: So you want to remind yourself of what or learn right now, what does the top court decay to what is it who is on decay to what is this he posts on what is the drill jaan process, these are fairly common backgrounds, that we. 128 00:22:07.350 --> 00:22:16.800 Matthew Bellis: That we deal with, so a top court decays at the fireman level to a w plus and a B, if this is a tough work. 129 00:22:17.880 --> 00:22:27.420 Matthew Bellis: It does not always have to be a B, but the chance of the Cork, not being a B is very, very small, and so, most of the time when you think about this process. 130 00:22:28.380 --> 00:22:46.920 Matthew Bellis: You think about a top going to a w and a B and then the w decay, is one of two ways the w decays to either a charge laptop and a neutrino or to a corporate a Cork, which we observe as a jet and a jet in the detector. 131 00:22:48.360 --> 00:22:53.460 Matthew Bellis: This is referred to as the electronic decay and this is the hydroponics okay. 132 00:22:55.050 --> 00:23:00.030 Matthew Bellis: And so we really think of our detector we really think of the detector as being. 133 00:23:01.530 --> 00:23:16.830 Matthew Bellis: What we observe is a jet that we can tag as a bee that we can identify that has a displaced vertex and then a charge laptop and some missing energy from the neutrino or three jets in which one of them has been identified as a deja. 134 00:23:18.420 --> 00:23:28.860 Matthew Bellis: Most of the time, if there is a top producer, there is also an Anti top Cork, there is single tops that are produced through the the cave and off Shell w. 135 00:23:30.300 --> 00:23:45.570 Matthew Bellis: But to first order you can think about there's always an Anti top when there was a top, and so this manifests in the detector as to be checks to charge leptons and to neutrinos these are opposite be charged which i'll point out now. 136 00:23:47.190 --> 00:23:54.210 Matthew Bellis: To be jack's to non be jetson in the charts of time and trina or to be jets and for other jets. 137 00:23:54.690 --> 00:24:06.990 Matthew Bellis: This down here is that had chronic decay, this is referred to as the electronic decay and this would be the semi electronic decay, where one of the top sophisticated hydroponically in their other has decayed semi electronically. 138 00:24:08.700 --> 00:24:13.770 Matthew Bellis: And so right off the BAT you may not see anything that immediately maps to. 139 00:24:15.240 --> 00:24:25.470 Matthew Bellis: to your ultimate on decay, but I will point out that inside the jets are charged leptons you can have pions that are produced that decade to a. 140 00:24:26.550 --> 00:24:39.840 Matthew Bellis: decade to a new on, you can have other charged particles in the JET that decay electronically, and so you can have charged leptons inside your jet Cone. 141 00:24:40.620 --> 00:24:55.320 Matthew Bellis: Those are generally identified as part of the JET So while we won't go through the details, for your analysis, you may want to look for what I refer to as isolated leptons leptons are that are removed and not in the Cone that you have to find this your chat. 142 00:24:56.610 --> 00:25:02.490 Matthew Bellis: But those are details for further down in your analysis, this is just first order thinking about what some of your backgrounds might be. 143 00:25:05.220 --> 00:25:24.030 Matthew Bellis: If you want to think about w or Z both on the case, the w can decay, it looks like I have duplicated here I don't know why I copied this twice, but the w will decay two charts leptin and the neutrino or two jet's in such a way that charges conserved Z decays to a. 144 00:25:25.710 --> 00:25:37.410 Matthew Bellis: Opposite the charts laptop charts leptons or neutrinos which are missing, energy or a couple of objects and you'll see a number of samples where you have w plus and w minus. 145 00:25:38.340 --> 00:25:42.510 Matthew Bellis: I can't off the top of my head I can't picture, right now, exactly what the production processes. 146 00:25:42.990 --> 00:25:47.430 Matthew Bellis: But we generally have a number of events where you have two w's that are opposite, we charge. 147 00:25:47.880 --> 00:25:55.650 Matthew Bellis: In which case you can get two opposite, we charged leptons a couple of neutrinos or jet charged slept on the neutrino for chats. 148 00:25:55.980 --> 00:26:11.160 Matthew Bellis: As the events you can write it out, and then I will admit I ran out of steam writing down all the possible combinations for wc events, but the wc can be a significant background process and I point out that there is. 149 00:26:12.510 --> 00:26:22.470 Matthew Bellis: You can have two charts and charge them to same sign leptons when you have w's and z's same thing with diseases, right here. 150 00:26:23.790 --> 00:26:33.150 Matthew Bellis: Now, in these processes, where you have WW zz you're not necessarily going to get only you know, two w's are only two z's. 151 00:26:33.900 --> 00:26:46.320 Matthew Bellis: You can have to see the effects where you know your corpse are the glue ons in the initial state may radiate and so, even when these types of events, you can have jets that are observed. 152 00:26:46.980 --> 00:26:57.240 Matthew Bellis: And so, just because i've written out for leptons here does not mean that you will see only for leptons and the data right now sorry in the simulation. 153 00:26:59.610 --> 00:27:19.320 Matthew Bellis: The dry Leon process is similar it's an interaction, where you have to corks that interact and couple to an off Shell photon or an off shelves Z which then decayed to two opposite sign leptons to charge leptons. 154 00:27:20.910 --> 00:27:38.400 Matthew Bellis: This is just an example of the cross section measured and predicted for the electronic new on channels and so you get a lot of opposite sign leptons you get a peek at the Z and then notice here the scale changes with between two and 600. 155 00:27:39.450 --> 00:27:46.740 Matthew Bellis: So you will see a number of samples designated as the why they generally produced opposite side. 156 00:27:48.480 --> 00:27:58.560 Matthew Bellis: And so, trying to figure out what the standard model samples are now i've moved away from the trigger and i'm just thinking about my analysis, ideally, the top and the anti top are mostly rejected. 157 00:27:58.950 --> 00:28:07.620 Matthew Bellis: If you require same sign leptons although you could get leptons from jack said it's up to you to make sure that these are isolated leptons. 158 00:28:08.130 --> 00:28:14.100 Matthew Bellis: The w plus w minus is generally rejected by saying time leptons maybe but it's always good to check. 159 00:28:14.610 --> 00:28:28.410 Matthew Bellis: But your wi fi and the zz could be background you're drilling you on could be rejected by same sign leptons This does not guarantee that there is no background from your top and anti top, this is a fairly large cross section. 160 00:28:29.520 --> 00:28:39.840 Matthew Bellis: And so you know, it is worth always trying to study, some of the larger backgrounds, even if you don't think that they're immediately going to contribute. 161 00:28:40.230 --> 00:28:49.530 Matthew Bellis: A large backgrounds, that you can reduce by 99% that 1% still might be big enough that it's a real real concern. 162 00:28:50.520 --> 00:29:00.360 Matthew Bellis: And if you go through if you go to the standard model simulation simulation samples the ar D Sim, you will find quite a few of these. 163 00:29:00.870 --> 00:29:07.650 Matthew Bellis: In there, for example, if you go to the standard model electroweak samples i've provided links to this. 164 00:29:08.400 --> 00:29:26.430 Matthew Bellis: You will see a number of things that are designated as wc or zz you'll see some WW stare and I have provided links, if you want to click on them and learn more so, this is a simulated sample of zz to two electrons to milan's so it has been. 165 00:29:28.500 --> 00:29:48.060 Matthew Bellis: constructed in such a way that it requires one of disease to decay to e plus C minus the other Z to decay to view plus minus, which brings me to another point about these samples let's see maybe I guess I didn't actually write any texts, but let me mention it here today. 166 00:29:49.380 --> 00:29:50.670 Matthew Bellis: hi yeah okay so i'm going to go back. 167 00:29:51.780 --> 00:29:57.360 Matthew Bellis: I will mention this on the next one about what this to eat you means. 168 00:29:58.380 --> 00:30:09.450 Matthew Bellis: I gave you a I gave you a link to a very specific sample to think about what that sample is and what it, whether or not it could be a significant background. 169 00:30:10.410 --> 00:30:24.210 Matthew Bellis: it's a top plus anti top plus jet so this TT jax means that there's a top and an Anti top but there's also jets that come from QC processes so, for example, radiative glue ons that then have an eyes. 170 00:30:26.160 --> 00:30:39.030 Matthew Bellis: from either the initial or final state the top and anti topic case me electronically, which means that one of them will decay hydroponically so you will still get three jets from one of them, and then the jets. 171 00:30:39.570 --> 00:30:54.690 Matthew Bellis: The charts laptop and the neutrino from the other, so this doesn't immediately seem like it could be contributing background, however, if if you are unlucky there may be a laptop from one of these jets that does not get. 172 00:30:55.980 --> 00:31:01.020 Matthew Bellis: That actually decays in such a way that it's outside the Cone so while it's unlikely. 173 00:31:01.620 --> 00:31:14.010 Matthew Bellis: Because the top anti top is a fairly large cross section, you may find yourself running studies to see how many past the trigger or your initial cups again, this is just to emphasize. 174 00:31:14.490 --> 00:31:24.000 Matthew Bellis: There is no one size fits all at this workshop, we cannot teach you how to do a full PhD quality level analysis but we'd like to point out some of the gotchas. 175 00:31:25.110 --> 00:31:31.350 Matthew Bellis: The other thing I wanted to mention is just like how the triggers there's an issue with double counting. 176 00:31:32.400 --> 00:31:42.690 Matthew Bellis: there's also a potential issue with double counting in the simulated data sets the fact that this TT jets you've required these semi electronic. 177 00:31:43.800 --> 00:31:45.390 Matthew Bellis: like mad graph the case. 178 00:31:47.340 --> 00:31:56.100 Matthew Bellis: If you have a TT jets sample where it has not required the semi electronic decay is, you will actually have this data set. 179 00:31:57.270 --> 00:32:08.010 Matthew Bellis: In that larger and that larger data sample, and this is because someone may want to study very, very specifically, the semi electronic the case with a larger. 180 00:32:08.340 --> 00:32:17.850 Matthew Bellis: amounts of statistics, but it means that if you're trying to do some acceptance correction some sort of efficiency, study and you're using. 181 00:32:19.080 --> 00:32:33.510 Matthew Bellis: These different data sets you've got to be careful again not to double count, that this is a subset of some other other data set but it's there for people who want a more targeted study of particular processes. 182 00:32:37.440 --> 00:32:47.370 Matthew Bellis: So that is a those are the solutions they are now posted there was a question about fake towels from TT and Q CD. 183 00:32:48.060 --> 00:33:02.100 Matthew Bellis: yeah that's a that's a great question Giles if there are if you're doing some study with towels right a towels kind of looks like a jet except for very specific case it's a cluster of energy usually with the particles, that is an issue. 184 00:33:04.020 --> 00:33:14.340 Matthew Bellis: There are real towels from top court case right if the top work decays electronically, you will get a real TAO in there and that could be mistaken for eject. 185 00:33:14.730 --> 00:33:26.040 Matthew Bellis: And so, as Julie was alluding to yesterday in many of the physics objects, you have these knobs that you can turn, is it like a tight identification, am I really, really sure that it's a towel. 186 00:33:27.000 --> 00:33:34.650 Matthew Bellis: You may have to crank that down, but you might throw out some good towels as well, this will affect your sensitivity, perhaps to your signal. 187 00:33:36.660 --> 00:33:37.800 Matthew Bellis: So it's a good question. 188 00:33:42.030 --> 00:33:58.020 Matthew Bellis: Any other questions, we can always continue this if you have any other questions that pop into your head in the next five minutes 10 minutes 10 days you can use the matter most channel for the data set scouting to post those questions again these slides are now posted on. 189 00:33:59.220 --> 00:34:01.530 Matthew Bellis: On the on the conference website. 190 00:34:02.700 --> 00:34:15.990 Matthew Bellis: So you can review them just out of curiosity there's no shame if you don't raise your hand i'm just curious, this is totally totally fine how many people spend some time trying to go through some of these. 191 00:34:17.100 --> 00:34:20.130 Matthew Bellis: exercises yesterday, thank you. 192 00:34:21.540 --> 00:34:33.240 Matthew Bellis: Thanks oh my gosh that's awesome anything more than one I would have been happy with so that's great oh my gosh look at all these hands i'm so i'm so excited to see that people were trying this out. 193 00:34:33.840 --> 00:34:48.030 Matthew Bellis: So many hands, I actually thought it was going to be like zero or one, not that I don't believe in you, but it's just a tremendous amount of work, so I really appreciate that the end of a long day people spend some time going through this. 194 00:34:49.410 --> 00:34:53.280 Matthew Bellis: Okay, so I think i've run to the end of my time. 195 00:34:54.450 --> 00:35:03.900 Matthew Bellis: i'm going to hand it over to Julie again if there are any questions that come up please post them in the matter most channel for this exercise. 196 00:35:05.010 --> 00:35:06.540 Matthew Bellis: Thank you, everybody for your engagement I. 197 00:35:06.540 --> 00:35:23.760 Matthew Bellis: will be seeing you online over the course of 10 tomorrow i'm going to hand it over to Julie Julie i'm going to stop this recording and then started again just before you start okay great alright, thank you, everybody bye bye good night good like. 198 00:35:24.090 --> 00:35:25.230 Matthew Bellis: Stop the cloud recording.