More actions
1 September 2024
- 16:0416:04, 1 September 2024 diff hist +5 Internal:FromGitlab/Bayes and Certainty No edit summary current
- 16:0416:04, 1 September 2024 diff hist +11,425 N Bayes and certainty Created page with "{{Main|Technical overview of the ratings system}} <span id="introduction--bayes-pulls-toward-certainty"></span> == Introduction – Bayes pulls toward certainty == We’ve already discussed the fact that Bayes pulls in favor of certainty. If we combine a 99% opinion with a 10% opinion we get 91.7%. But if we increase the 99% to 99.9% the combined opinion rises to 99.1%. If we increase yet again to 99.99% the combined opinion rises to 99.91%. We summarize this below to..."
- 16:0416:04, 1 September 2024 diff hist +110 Technical overview of the ratings system No edit summary
- 15:5615:56, 1 September 2024 diff hist +51 Error bars and a problem with Bayesian modeling No edit summary
- 15:5515:55, 1 September 2024 diff hist +5 Internal:FromGitlab/Error Bars and a Problem with Bayesian Modeling No edit summary current
- 15:5515:55, 1 September 2024 diff hist +4,203 N Error bars and a problem with Bayesian modeling Created page with "<h2>Error Bars for Incomplete Analyses</h2> During last week’s meeting (6/19/23) Eric suggested the idea of putting error bars around incomplete results as they stream in. To do this we can compute the Pcomb for all the nodes in the network, whether they’ve answered or not, by putting P=50% on the nodes that haven’t answered. This way they don’t affect the calculation and the calculation can remain as simple as possible: just compute as if you have all the infor..."
- 15:5415:54, 1 September 2024 diff hist +158 Technical overview of the ratings system No edit summary
- 15:4915:49, 1 September 2024 diff hist +20 Establish trust No edit summary
- 15:2615:26, 1 September 2024 diff hist +5 Internal:FromGitlab/Ideas for Establishing Trust No edit summary current
- 15:2215:22, 1 September 2024 diff hist −4 Trust No edit summary
- 15:2115:21, 1 September 2024 diff hist +5 Internal:FromGitlab/Modification to the Sapienza probability adjustment for trust to include random lying, bias, and biased lying No edit summary current
- 15:2115:21, 1 September 2024 diff hist +8,837 N Modification to the Sapienza probability adjustment for trust to include random lying, bias, and biased lying Created page with "{{Main|Trust}} In the last iteration on this subject we modified the probability adjustment for trust to include lying and bias. We noted that the lying was random in nature, that is, the lie would be distributed evenly among all options that were not the truth. Here we modify this assumption by allowing, in addition, lying that is biased toward a particular o..."
- 15:1915:19, 1 September 2024 diff hist +6 Internal:FromGitlab/Modification to the Sapienza probability adjustment for trust to include lying and bias No edit summary current
- 15:1815:18, 1 September 2024 diff hist 0 Modification to the Sapienza probability adjustment for trust to include lying and bias No edit summary
- 15:1715:17, 1 September 2024 diff hist +16 Modification to the Sapienza probability adjustment for trust to include lying and bias No edit summary
- 15:1615:16, 1 September 2024 diff hist +9,835 N Modification to the Sapienza probability adjustment for trust to include lying and bias Created page with "<h2>Review of Sapienza's Probability Adjustment Equation</h2> Sapienza’s paper (https://ceur-ws.org/Vol-1664/w9.pdf), which is the basis for our modeling approach, uses the notion of trust between nodes to adjust the node’s probability of getting a particular answer. Each node is first assigned probabilities for a particular outcome (eg 60% Red, 30% Blue, 10% Green). Then, based on trust, the nodes’ probabilities are adjusted (or smoothed, as Sapienza puts it) up..."
- 15:1415:14, 1 September 2024 diff hist −4 Trust No edit summary
- 15:1415:14, 1 September 2024 diff hist −4 Trust No edit summary
- 15:1415:14, 1 September 2024 diff hist +23 Trust No edit summary
- 15:0515:05, 1 September 2024 diff hist +5 Internal:FromGitlab/Calculations for the two trust factor algorithm No edit summary current
- 15:0415:04, 1 September 2024 diff hist +32 Trust attenuation and the inadequacy of single-value trust factors No edit summary
- 15:0315:03, 1 September 2024 diff hist +6,643 Trust attenuation and the inadequacy of single-value trust factors No edit summary
- 14:5714:57, 1 September 2024 diff hist −56 Trust attenuation and the inadequacy of single-value trust factors No edit summary
- 14:5514:55, 1 September 2024 diff hist +51 Trust attenuation and the inadequacy of single-value trust factors No edit summary
- 14:5414:54, 1 September 2024 diff hist +5 Internal:FromGitlab/Followup on Trust Attenuation and the Inadequacy of Single Value Trust Factors No edit summary current
- 14:5314:53, 1 September 2024 diff hist +7,773 N Trust attenuation and the inadequacy of single-value trust factors Created page with "During last week’s meeting (6/7/23) we discussed, among other things, trust attenuation: Attenuation in Trust Networks. Let’s start by following up on some points that were raised about this: ==== Why we're not using a tree topology ==== The examples show a single straight line representing the originating client (top-node) to the server (leaf node) that answers the query. The rest of the tree is simply not show..."
- 14:5214:52, 1 September 2024 diff hist +298 Technical overview of the ratings system No edit summary
- 14:4514:45, 1 September 2024 diff hist −180 Attenuation in trust networks No edit summary
- 14:4314:43, 1 September 2024 diff hist +51 Attenuation in trust networks No edit summary
- 14:4214:42, 1 September 2024 diff hist +5 Internal:FromGitlab/Attenuation in Trust Networks No edit summary current
- 14:4114:41, 1 September 2024 diff hist +11,093 N Attenuation in trust networks Created page with "<h2>Review of Cycling</h2> During last week’s call we discussed, among other things, the effect of cycling and how it can badly distort confidence results: Effect of Cycling in Trust Networks. To summarize briefly, if Trust=1.0 for all nodes, then cycling will rapidly lead to a confidence of 1 (100%) for a given question (if probabilities for most of the nodes are above 50%). However, if Trust < 1.0 then t..."
- 14:4014:40, 1 September 2024 diff hist +49 Technical overview of the ratings system No edit summary
- 14:3714:37, 1 September 2024 diff hist +19 Technical overview of the ratings system No edit summary
- 14:3314:33, 1 September 2024 diff hist +5 Internal:FromGitlab/Effect of Cycling in Trust Networks No edit summary current
- 14:3214:32, 1 September 2024 diff hist +51 Effect of cycling in trust networks No edit summary
- 14:3114:31, 1 September 2024 diff hist +4,555 N Effect of cycling in trust networks Created page with "This discussion will show how cycling in trust networks rapidly leads to an incorrect answer (Pcomb) which is often much higher than the answer you would get in a non-cycling network. The divergence between answers, however, depends on trust. If trust is low the answers will tend to converge and will be equal when trust is zero. If trust is high the answers will be very far apart. Suppose we have a trust network composed of four Nodes 0, 1, 2, 3 with the following conne..."
- 14:3014:30, 1 September 2024 diff hist +40 Technical overview of the ratings system No edit summary
31 August 2024
- 15:3515:35, 31 August 2024 diff hist +78 Technical overview of the ratings system No edit summary
- 15:3015:30, 31 August 2024 diff hist +559 Technical overview of the ratings system No edit summary
30 August 2024
- 21:4321:43, 30 August 2024 diff hist +119 Aggregation techniques No edit summary
- 21:3721:37, 30 August 2024 diff hist −35 Trust/Probability/Population graphs algorithm No edit summary
- 21:3621:36, 30 August 2024 diff hist −44 Trust/Probability/Population graphs algorithm No edit summary
- 21:3521:35, 30 August 2024 diff hist +33 Trust/Probability/Population graphs algorithm No edit summary
- 21:3421:34, 30 August 2024 diff hist +12 Internal:FromGitlab/More algorithms for the interface No edit summary current
- 21:3421:34, 30 August 2024 diff hist +9,823 N Trust/Probability/Population graphs algorithm Created page with "== Trust/Probability/Population Graphs Algorithm U99== This algorithm is a variant of Eric’s trust-weighted histogram (TWH) algorithm and can be conceived of as follows. Start by plotting probability on the x-axis and trust on the y-axis for each source. Now draw lines to represent intervals in the x and y-axis (eg 0-0.2, 0.2-0.4, etc). Each rectangle (ie bin) in the resulting grid will contain some..." Tag: Visual edit: Switched
- 21:3221:32, 30 August 2024 diff hist +98 Aggregation techniques No edit summary
- 21:2921:29, 30 August 2024 diff hist +4 Internal:FromGitlab/More algorithms for the interface No edit summary
- 21:2721:27, 30 August 2024 diff hist +36 Allowing for more than predicate questions in the trust-weighted histogram (TWH) algorithm No edit summary
- 21:2621:26, 30 August 2024 diff hist +5,954 N Allowing for more than predicate questions in the trust-weighted histogram (TWH) algorithm Created page with "The <code>trust_weighted_histogram</code> (TWH) algorithm was discussed last time but only handles predicate questions. Indeed, it only handles single valued probabilities under the assumption that the other probability in a predicate distribution is (1-P). Although the examples that exercise it are actually provided as a two-valued distribution [P, 1-P], the algorithm ignores the 2nd value. This was corrected using the <code>trust_weighted_histogram_sets</code> algorit..."
- 21:2521:25, 30 August 2024 diff hist +179 Trust-weighted histograms No edit summary