<div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div>Dear All,<div><div dir="ltr"><div style="font-family:inherit;background-color:transparent;background-image:inherit;background-position:inherit;background-size:inherit;background-repeat:inherit;background-origin:inherit;background-clip:inherit;border-right:inherit;font-size:1.375rem;line-height:28px;margin:0px;padding:0px"><span style="background-color:transparent;font-size:small;font-family:inherit">You are cordially invited to the  Weekly Online Seminar “Analysis and Applied Mathematics” on  </span><span style="background-color:transparent;font-family:inherit;font-size:1.375rem">  </span></div><div style="font-family:inherit;background-color:transparent;background-image:inherit;background-position:inherit;background-size:inherit;background-repeat:inherit;background-origin:inherit;background-clip:inherit;border-right:inherit;font-size:1.375rem;line-height:28px;margin:0px;padding:0px"><b style="font-size:small">Date</b><span style="font-size:small">: Tuesday, February 0</span><span style="font-size:small">4, 2025 </span></div></div><div dir="ltr"><b>Time:</b> 14.00-15.00 (Istanbul) = 13.00-14.00 (Ghent) = 16.00-17.00 (Almaty)<br><b>Place:</b> Zoom link: <a href="https://us02web.zoom.us/j/6678270445?pwd=SFNmQUIvT0tRaHlDaVYrN3l5bzJVQT09" rel="noreferrer" target="_blank">https://us02web.zoom.us/j/6678270445?pwd=SFNmQUIvT0tRaHlDaVYrN3l5bzJVQT09</a>, Conference ID: 667 827 0445, Access code: 1  </div><div dir="ltr"> </div><b>Speaker:  Assoc. Prof. Dr. Betül Hiçdurmaz</b> </div><div>Istanbul Medeniyet University, Istanbul, Türkiye</div><div><br></div><div><b>Title: </b>

Completely Monotonic Functions: Driving Innovation in
Machine Learning Models.  </div><div><b>Abstract:</b> 

Completely monotonic functions have long been a subject of mathematical and
theoretical interest due to their unique properties, including their role in constructing valid
kernel functions in machine learning. This talk delves into the mathematical theory behind
the Logarithmic Kernel Function (LKF), a novel kernel designed to enhance machine learning models in scenarios with limited training data. LKF is inspired by the well-known radial
basis function (RBF) kernel, sharing a Gaussian-like structure while offering superior performance under specific conditions. By leveraging the Mercer condition, this kernel broadens the boundaries of classification and regression tasks, as demonstrated through rigorous
experimentation. Using datasets across various domains, the LKF demonstrates its ability
to outperform traditional kernels, such as RBF, especially when combined with advanced
hyperparameter optimization techniques like the Tree Parzen Estimator.
The talk will also explore the broader implications of completely monotonic functions as
kernel candidates within the framework of machine learning theory. These functions provide an avenue for creating innovative kernels with tailored properties, expanding the versatility and adaptability of models to non-linear and high-dimensional data. By bridging
mathematical rigor and practical application, this discussion aims to shed light on the potential of completely monotone functions in shaping the future of kernel-based learning algorithms.

</div><div><div><div dir="ltr"> <b>Abstracts and forthcoming talks can be found on our webpage</b></div><div dir="ltr"><div><div><div><div><div><a href="https://sites.google.com/view/aam-seminars" rel="noreferrer noreferrer" target="_blank">https://sites.google.com/view/aam-seminars</a><br>With my best wishes</div></div></div></div></div></div></div></div></div><div><div dir="ltr" class="gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div style="font-size:12.8px"><b style="font-size:12.8px">Prof. Dr. Allaberen Ashyralyev </b></div><div><span style="font-size:small"><b>Department of Mathematics, Bahcesehir University,</b></span><b>34349</b><span style="font-size:small"><b>, Istanbul, Turkiye</b></span><b style="font-size:12.8px">  </b></div><div><div style="font-size:12.8px"><span style="font-size:12.8px"><b>Peoples' Friendship University of Russia (RUDN University),</b></span><b style="color:rgb(51,51,51);font-family:arial,helvetica,sans-serif;font-size:13px"> Ul Miklukho Maklaya 6, Moscow 117198, Russian Federation </b></div><div style="font-size:12.8px"><b style="font-size:12.8px">Institute of Mathematics and Mathematical Modelling, 050010, Almaty, Kazakhstan</b></div><div style="font-size:12.8px"><b style="font-size:12.8px">e-mail: <a href="mailto:allaberen.ashyralyev@neu.edu.tr" style="color:rgb(17,85,204)" target="_blank">allaberen.ashyralyev@bau.edu.tr</a> and </b><b style="font-size:12.8px"><a href="mailto:aallaberen@gmail.com" style="color:rgb(17,85,204)" target="_blank">aallaberen@gmail.com</a> </b></div><div style="font-size:12.8px"><b style="font-size:12.8px"> </b><a href="http://akademik.bahcesehir.edu.tr/web/allaberenashyralyev" style="color:rgb(0,86,179);margin:0px;padding:0px;border:0px;font-stretch:inherit;font-size:14px;line-height:inherit;font-family:Roboto,sans-serif;vertical-align:baseline" target="_blank">http://akademik.bahcesehir.edu.tr/web/allaberenashyralyev</a></div><div><p class="MsoNormal"><a href="https://sites.google.com/view/aam-seminars" target="_blank">https://sites.google.com/view/aam-seminars</a></p><p class="MsoNormal"><a href="https://ejaam.org/editorial.html" target="_blank">https://ejaam.org/editorial.html</a><br></p><p class="MsoNormal"><span style="font-size:12.8px"><b><u><a href="https://icaam-online.org/" target="_blank">https://icaam-online.org/</a></u></b></span><br></p></div></div><div><font color="#1155cc"><span style="font-size:12.8px"><b><a href="https://www.genealogy.math.ndsu.nodak.edu/id.php?id=95872&fChrono=1" target="_blank">https://www.genealogy.math.ndsu.nodak.edu/id.php?id=95872&fChrono=1<br></a></b></span><br></font></div></div></div></div></div></div></div></div></div></div></div></div></div></div>
</div>
</div>
</div>