https://sdq.kastel.kit.edu/index.php?title=Patient_Rule_Induction_Method_with_Active_Learning&feed=atom&action=history
Patient Rule Induction Method with Active Learning - Versionsgeschichte
2024-03-29T08:59:45Z
Versionsgeschichte dieser Seite in SDQ-Institutsseminar
MediaWiki 1.39.6
https://sdq.kastel.kit.edu/mediawiki-institutsseminar/index.php?title=Patient_Rule_Induction_Method_with_Active_Learning&diff=1231&oldid=prev
Ubesb am 26. November 2019 um 17:44 Uhr
2019-11-26T17:44:15Z
<p></p>
<table style="background-color: #fff; color: #202122;" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="de">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Nächstältere Version</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Version vom 26. November 2019, 18:44 Uhr</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l5">Zeile 5:</td>
<td colspan="2" class="diff-lineno">Zeile 5:</td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>|betreuer=Vadim Arzamasov</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>|betreuer=Vadim Arzamasov</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>|termin=Institutsseminar/2019-11-29 Zusatztermin</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>|termin=Institutsseminar/2019-11-29 Zusatztermin</div></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>|kurzfassung=PRIM (Patient Rule Induction Method) is an algorithm for discovering scenarios from simulations, by creating hyperboxes, that are human-comprehensible. Yet PRIM alone requires relatively large datasets and computational simulations are usually quite expensive. <del style="font-weight: bold; text-decoration: none;">Therefore</del>, one wants to obtain a plausible scenario, with a minimal number of simulations. It has been shown, that combining PRIM with ML models, which generalize faster, can reduce the number of necessary simulation runs by around 75%.</div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>|kurzfassung=PRIM (Patient Rule Induction Method) is an algorithm for discovering scenarios from simulations, by creating hyperboxes, that are human-comprehensible. Yet PRIM alone requires relatively large datasets and computational simulations are usually quite expensive. <ins style="font-weight: bold; text-decoration: none;">Consequently</ins>, one wants to obtain a plausible scenario, with a minimal number of simulations. It has been shown, that combining PRIM with ML models, which generalize faster, can reduce the number of necessary simulation runs by around 75%.</div></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>We will try to reduce the number of simulation runs even further, using an active learning approach to train <del style="font-weight: bold; text-decoration: none;">a suitable </del>model. </div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>We will try to reduce the number of simulation runs even further, using an active learning approach to train <ins style="font-weight: bold; text-decoration: none;">an intermediate ML </ins>model. </div></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div><del style="font-weight: bold; text-decoration: none;">Moreover</del>, we <del style="font-weight: bold; text-decoration: none;">are interested in applying PRIM with Active Learning </del>to classification <del style="font-weight: bold; text-decoration: none;">as well as </del>regression problems<del style="font-weight: bold; text-decoration: none;">, utilizing different query strategies. Acquiring labels for a given dataset can be quite costly, with active learning, only a small part of the dataset has to be labeled</del>. A preliminary experiment indicated, that the combination of these methods, does indeed help reduce the necessary runs even further. <del style="font-weight: bold; text-decoration: none;">Though we only tried one possible </del>sampling method <del style="font-weight: bold; text-decoration: none;">thus far. Finding the best-suited combinations of components </del>for <del style="font-weight: bold; text-decoration: none;">the above described non-trivial problems is the focus of </del>this <del style="font-weight: bold; text-decoration: none;">Thesis</del>.</div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;">Additionally</ins>, we <ins style="font-weight: bold; text-decoration: none;">extend the previously proposed methodology </ins>to <ins style="font-weight: bold; text-decoration: none;">not only cover </ins>classification <ins style="font-weight: bold; text-decoration: none;">but also </ins>regression problems. A preliminary experiment indicated, that the combination of these methods, does indeed help reduce the necessary runs even further. <ins style="font-weight: bold; text-decoration: none;">In this thesis, I will analyze different AL sampling strategies together with several intermediate ML models to find out if AL can systematically improve existing scenario discovery methods and if a most beneficial combination of </ins>sampling method <ins style="font-weight: bold; text-decoration: none;">and intermediate ML model exists </ins>for this <ins style="font-weight: bold; text-decoration: none;">purpose</ins>.</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>}}</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>}}</div></td></tr>
</table>
Ubesb
https://sdq.kastel.kit.edu/mediawiki-institutsseminar/index.php?title=Patient_Rule_Induction_Method_with_Active_Learning&diff=1230&oldid=prev
Ubesb am 26. November 2019 um 14:55 Uhr
2019-11-26T14:55:50Z
<p></p>
<table style="background-color: #fff; color: #202122;" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="de">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Nächstältere Version</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Version vom 26. November 2019, 15:55 Uhr</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l5">Zeile 5:</td>
<td colspan="2" class="diff-lineno">Zeile 5:</td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>|betreuer=Vadim Arzamasov</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>|betreuer=Vadim Arzamasov</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>|termin=Institutsseminar/2019-11-29 Zusatztermin</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>|termin=Institutsseminar/2019-11-29 Zusatztermin</div></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>|kurzfassung=PRIM (Patient Rule Induction Method) is an algorithm <del style="font-weight: bold; text-decoration: none;">to create </del>hyperboxes, that are human-comprehensible. Yet PRIM alone requires relatively large datasets. It has been shown, that combining PRIM with ML models <del style="font-weight: bold; text-decoration: none;">(e.g. Random Forrest)</del>, which generalize faster, can reduce the number of simulation runs by around 75%.</div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>|kurzfassung=PRIM (Patient Rule Induction Method) is an algorithm <ins style="font-weight: bold; text-decoration: none;">for discovering scenarios from simulations, by creating </ins>hyperboxes, that are human-comprehensible. Yet PRIM alone requires relatively large datasets <ins style="font-weight: bold; text-decoration: none;">and computational simulations are usually quite expensive. Therefore, one wants to obtain a plausible scenario, with a minimal number of simulations</ins>. It has been shown, that combining PRIM with ML models, which generalize faster, can reduce the number of <ins style="font-weight: bold; text-decoration: none;">necessary </ins>simulation runs by around 75%.</div></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>We <del style="font-weight: bold; text-decoration: none;">are trying </del>to reduce the number of simulation runs even further, using an active learning approach to train <del style="font-weight: bold; text-decoration: none;">the </del>model. </div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>We <ins style="font-weight: bold; text-decoration: none;">will try </ins>to reduce the number of simulation runs even further, using an active learning approach to train <ins style="font-weight: bold; text-decoration: none;">a suitable </ins>model. </div></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>Moreover, we are interested in applying <del style="font-weight: bold; text-decoration: none;">this approach </del>to classification as well as regression problems, utilizing different query strategies.</div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>Moreover, we are interested in applying <ins style="font-weight: bold; text-decoration: none;">PRIM with Active Learning </ins>to classification as well as regression problems, utilizing different query strategies. Acquiring labels for a given dataset can be quite costly, with active learning, only a small part of the dataset has to be labeled. A preliminary experiment indicated, that the combination of these methods, does indeed help reduce the necessary runs even further. Though we only tried one possible sampling method thus far. <ins style="font-weight: bold; text-decoration: none;">Finding the best-suited combinations of components for the above described </ins>non-trivial <ins style="font-weight: bold; text-decoration: none;">problems </ins>is the focus of this Thesis.</div></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>Acquiring labels for a given dataset can be quite costly, with active learning, only a small part of the dataset has to be labeled. A preliminary experiment indicated, that the combination of these methods, does indeed help reduce the necessary runs even further. Though we only tried one possible sampling method thus far. <del style="font-weight: bold; text-decoration: none;">Optimizing this </del>non-trivial <del style="font-weight: bold; text-decoration: none;">task </del>is the focus of this Thesis.</div></td><td colspan="2" class="diff-side-added"></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>}}</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>}}</div></td></tr>
</table>
Ubesb
https://sdq.kastel.kit.edu/mediawiki-institutsseminar/index.php?title=Patient_Rule_Induction_Method_with_Active_Learning&diff=1229&oldid=prev
Ubesb am 26. November 2019 um 09:30 Uhr
2019-11-26T09:30:47Z
<p></p>
<table style="background-color: #fff; color: #202122;" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="de">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Nächstältere Version</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Version vom 26. November 2019, 10:30 Uhr</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l5">Zeile 5:</td>
<td colspan="2" class="diff-lineno">Zeile 5:</td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>|betreuer=Vadim Arzamasov</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>|betreuer=Vadim Arzamasov</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>|termin=Institutsseminar/2019-11-29 Zusatztermin</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>|termin=Institutsseminar/2019-11-29 Zusatztermin</div></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>|kurzfassung=PRIM (Patient Rule Induction Method) is an algorithm to create hyperboxes that are human <del style="font-weight: bold; text-decoration: none;">comprehansable</del>. <del style="font-weight: bold; text-decoration: none;">But </del>PRIM requires relatively large datasets. It has been shown, that <del style="font-weight: bold; text-decoration: none;">using </del>ML models (e.g. Random Forrest) <del style="font-weight: bold; text-decoration: none;">that </del>generalize faster can <del style="font-weight: bold; text-decoration: none;">increase performance </del>by around 75%.</div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>|kurzfassung=PRIM (Patient Rule Induction Method) is an algorithm to create hyperboxes<ins style="font-weight: bold; text-decoration: none;">, </ins>that are human<ins style="font-weight: bold; text-decoration: none;">-comprehensible</ins>. <ins style="font-weight: bold; text-decoration: none;">Yet </ins>PRIM <ins style="font-weight: bold; text-decoration: none;">alone </ins>requires relatively large datasets. It has been shown, that <ins style="font-weight: bold; text-decoration: none;">combining PRIM with </ins>ML models (e.g. Random Forrest)<ins style="font-weight: bold; text-decoration: none;">, which </ins>generalize faster<ins style="font-weight: bold; text-decoration: none;">, </ins>can <ins style="font-weight: bold; text-decoration: none;">reduce the number of simulation runs </ins>by around 75%.</div></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div><del style="font-weight: bold; text-decoration: none;">In this Thesis we </del>are trying to <del style="font-weight: bold; text-decoration: none;">increase </del>the <del style="font-weight: bold; text-decoration: none;">overall performance </del>even further, using an active learning approach <del style="font-weight: bold; text-decoration: none;">in order </del>to train the <del style="font-weight: bold; text-decoration: none;">models</del>. Acquiring labels for a given dataset can be quite costly, with active learning only a small part of the dataset has to <del style="font-weight: bold; text-decoration: none;">ben </del>labeled <del style="font-weight: bold; text-decoration: none;">(if at all)</del>. <del style="font-weight: bold; text-decoration: none;">Furthermore, a </del>preliminary experiment indicated, that <del style="font-weight: bold; text-decoration: none;">combining </del>these methods does indeed <del style="font-weight: bold; text-decoration: none;">increase performance </del>even further.</div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;">We </ins>are trying to <ins style="font-weight: bold; text-decoration: none;">reduce </ins>the <ins style="font-weight: bold; text-decoration: none;">number of simulation runs </ins>even further, using an active learning approach to train the <ins style="font-weight: bold; text-decoration: none;">model. </ins></div></td></tr>
<tr><td colspan="2" class="diff-side-deleted"></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;">Moreover, we are interested in applying this approach to classification as well as regression problems, utilizing different query strategies</ins>.</div></td></tr>
<tr><td colspan="2" class="diff-side-deleted"></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>Acquiring labels for a given dataset can be quite costly, with active learning<ins style="font-weight: bold; text-decoration: none;">, </ins>only a small part of the dataset has to <ins style="font-weight: bold; text-decoration: none;">be </ins>labeled. <ins style="font-weight: bold; text-decoration: none;">A </ins>preliminary experiment indicated, that <ins style="font-weight: bold; text-decoration: none;">the combination of </ins>these methods<ins style="font-weight: bold; text-decoration: none;">, </ins>does indeed <ins style="font-weight: bold; text-decoration: none;">help reduce the necessary runs </ins>even further<ins style="font-weight: bold; text-decoration: none;">. Though we only tried one possible sampling method thus far. Optimizing this non-trivial task is the focus of this Thesis</ins>.</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>}}</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>}}</div></td></tr>
</table>
Ubesb
https://sdq.kastel.kit.edu/mediawiki-institutsseminar/index.php?title=Patient_Rule_Induction_Method_with_Active_Learning&diff=1228&oldid=prev
Nv3463 am 25. November 2019 um 15:43 Uhr
2019-11-25T15:43:06Z
<p></p>
<table style="background-color: #fff; color: #202122;" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="de">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Nächstältere Version</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Version vom 25. November 2019, 16:43 Uhr</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l4">Zeile 4:</td>
<td colspan="2" class="diff-lineno">Zeile 4:</td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>|vortragstyp=Proposal</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>|vortragstyp=Proposal</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>|betreuer=Vadim Arzamasov</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>|betreuer=Vadim Arzamasov</div></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>|termin=Institutsseminar/2019-11-29</div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>|termin=Institutsseminar/2019-11-29 <ins style="font-weight: bold; text-decoration: none;">Zusatztermin</ins></div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>|kurzfassung=PRIM (Patient Rule Induction Method) is an algorithm to create hyperboxes that are human comprehansable. But PRIM requires relatively large datasets. It has been shown, that using ML models (e.g. Random Forrest) that generalize faster can increase performance by around 75%.</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>|kurzfassung=PRIM (Patient Rule Induction Method) is an algorithm to create hyperboxes that are human comprehansable. But PRIM requires relatively large datasets. It has been shown, that using ML models (e.g. Random Forrest) that generalize faster can increase performance by around 75%.</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>In this Thesis we are trying to increase the overall performance even further, using an active learning approach in order to train the models. Acquiring labels for a given dataset can be quite costly, with active learning only a small part of the dataset has to ben labeled (if at all). Furthermore, a preliminary experiment indicated, that combining these methods does indeed increase performance even further.</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>In this Thesis we are trying to increase the overall performance even further, using an active learning approach in order to train the models. Acquiring labels for a given dataset can be quite costly, with active learning only a small part of the dataset has to ben labeled (if at all). Furthermore, a preliminary experiment indicated, that combining these methods does indeed increase performance even further.</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>}}</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>}}</div></td></tr>
</table>
Nv3463
https://sdq.kastel.kit.edu/mediawiki-institutsseminar/index.php?title=Patient_Rule_Induction_Method_with_Active_Learning&diff=1216&oldid=prev
Ubesb am 22. November 2019 um 11:35 Uhr
2019-11-22T11:35:02Z
<p></p>
<table style="background-color: #fff; color: #202122;" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="de">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Nächstältere Version</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Version vom 22. November 2019, 12:35 Uhr</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l5">Zeile 5:</td>
<td colspan="2" class="diff-lineno">Zeile 5:</td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>|betreuer=Vadim Arzamasov</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>|betreuer=Vadim Arzamasov</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>|termin=Institutsseminar/2019-11-29</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>|termin=Institutsseminar/2019-11-29</div></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>|kurzfassung=<del style="font-weight: bold; text-decoration: none;">Kurzfassung</del></div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>|kurzfassung=<ins style="font-weight: bold; text-decoration: none;">PRIM (Patient Rule Induction Method) is an algorithm to create hyperboxes that are human comprehansable. But PRIM requires relatively large datasets. It has been shown, that using ML models (e.g. Random Forrest) that generalize faster can increase performance by around 75%.</ins></div></td></tr>
<tr><td colspan="2" class="diff-side-deleted"></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;">In this Thesis we are trying to increase the overall performance even further, using an active learning approach in order to train the models. Acquiring labels for a given dataset can be quite costly, with active learning only a small part of the dataset has to ben labeled (if at all). Furthermore, a preliminary experiment indicated, that combining these methods does indeed increase performance even further.</ins></div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>}}</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>}}</div></td></tr>
</table>
Ubesb
https://sdq.kastel.kit.edu/mediawiki-institutsseminar/index.php?title=Patient_Rule_Induction_Method_with_Active_Learning&diff=1210&oldid=prev
Ubesb: Die Seite wurde neu angelegt: „{{Vortrag |vortragender=Emmanouil Emmanouilidis |email=ubesb@student.kit.edu |vortragstyp=Proposal |betreuer=Vadim Arzamasov |termin=Institutsseminar/2019-11-2…“
2019-11-19T15:54:07Z
<p>Die Seite wurde neu angelegt: „{{Vortrag |vortragender=Emmanouil Emmanouilidis |email=ubesb@student.kit.edu |vortragstyp=Proposal |betreuer=Vadim Arzamasov |termin=Institutsseminar/2019-11-2…“</p>
<p><b>Neue Seite</b></p><div>{{Vortrag<br />
|vortragender=Emmanouil Emmanouilidis<br />
|email=ubesb@student.kit.edu<br />
|vortragstyp=Proposal<br />
|betreuer=Vadim Arzamasov<br />
|termin=Institutsseminar/2019-11-29<br />
|kurzfassung=Kurzfassung<br />
}}</div>
Ubesb