<div dir="ltr"><br><div class="gmail_quote"><div class="gmail_attr">Caros redistas,</div><div class="gmail_attr"><br></div><div class="gmail_attr">Nesta semana teremos a presença de Luca Martino (URJC, Espanha) no seminário FGV/EMAp, falando sobre amostradores por importância e suas interações com MCMC. O seminário ocorre às quintas-feiras às 16h de Brasília. Mais informações abaixo.</div><div class="gmail_attr"><br></div><div class="gmail_attr">Esperamos vocês!</div><div class="gmail_attr"><br></div><div class="gmail_attr">Cordialmente,</div><div class="gmail_attr"><br></div><div class="gmail_attr">Luiz</div><br>





<div lang="PT-BR" link="#0563C1" vlink="#954F72">
<div class="m_-1617477441496007571WordSection1">
<p class="MsoNormal"><u></u> <u></u></p>
<table border="0" cellspacing="0" cellpadding="0" align="left" width="100%" style="width:100.0%;background:white;margin-left:4.8pt;margin-right:4.8pt">
<tbody>
<tr>
<td width="100%" style="width:100.0%;padding:0cm 0cm 0cm 0cm">
<table border="0" cellspacing="0" cellpadding="0" align="left" width="100%" style="width:100.0%;background:white;margin-left:4.8pt;margin-right:4.8pt">
<tbody>
<tr>
<td width="100%" style="width:100.0%;padding:0cm 0cm 0cm 0cm">
<table border="0" cellspacing="0" cellpadding="0" align="left" width="100%" style="width:100.0%;background:white;margin-left:4.8pt;margin-right:4.8pt">
<tbody>
<tr>
<td width="99%" colspan="2" style="width:99.96%;padding:0cm 0cm 0cm 0cm">
<table border="0" cellspacing="0" cellpadding="0" align="left" width="100%" style="width:100.0%;background:white;margin-left:4.8pt;margin-right:4.8pt">
<tbody>
<tr>
<td width="100%" colspan="3" style="width:100.0%;padding:0cm 0cm 0cm 0cm">
<p class="m_-1617477441496007571xmsonormal" align="center" style="text-align:center;line-height:105%">
<span style="font-size:13.5pt;line-height:105%;color:black"><img width="600" height="182" style="width:6.25in;height:1.9in" id="m_-1617477441496007571Imagem_x0020_1" src="cid:17b72e013b74ce8e91" alt="cid:image001.jpg@01D56E0B.2914C3E0"></span><u></u><u></u></p>
</td>
</tr>
<tr style="height:27.0pt">
<td width="0%" style="width:.06%;padding:0cm 0cm 0cm 0cm;height:27.0pt"></td>
<td width="99%" style="width:99.88%;padding:0cm 0cm 0cm 0cm;height:27.0pt">
<p class="MsoNormal">
<span style="font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1f497d"><u></u> <u></u></span></p>
<div align="center">
<table border="0" cellspacing="0" cellpadding="0" width="0" style="width:450.0pt;background:#fefefe">
<tbody>
<tr style="height:57.75pt">
<td width="600" colspan="3" style="width:450.0pt;padding:0cm 0cm 0cm 0cm;height:57.75pt">
<p class="MsoNormal" align="center" style="text-align:center;line-height:115%">
<b><span style="font-size:13.0pt;line-height:115%;font-family:"Arial",sans-serif;color:#2f5597">MCMC-DRIVEN IMPORTANCE SAMPLERS</span></b><b><span style="font-size:13.0pt;line-height:115%;font-family:"Arial",sans-serif"><u></u><u></u></span></b></p>
</td>
</tr>
<tr>
<td width="40" style="width:30.0pt;padding:0cm 0cm 0cm 0cm"></td>
<td width="520" style="width:390.0pt;padding:0cm 0cm 0cm 0cm">
<p class="MsoNormal" style="margin-bottom:12.0pt;text-align:justify;line-height:115%">
<span style="font-size:10.5pt;line-height:115%;font-family:"Arial",sans-serif;color:#595959">In this presentation, I focus on the layered adaptive importance sampling (LAIS), a class of adaptive importance samplers where Markov chain Monte Carlo (MCMC) algorithms
 are employed to drive an underlying multiple importance sampling (IS) scheme. Its power lies in the simplicity of the layered framework: the upper layer locates proposal densities by means of MCMC algorithms; while the lower layer handles the multiple IS scheme,
 in order to compute the final estimators. The modular nature of LAIS allows for different possible choices in the upper and lower layers, that will have different performance and computational costs. I show different enhancements in order to increase the efficiency
 and reduce the computational cost, of both upper and lower layers. The different variants are essential if we aim to address computational challenges arising in real-world applications, such as highly concentrated posterior distributions (due to large amounts
 of data, etc.). Hamiltonian-driven importance samplers will be presented and. Furthermore, I will describe different strategies for designing cheaper schemes, for instance, recycling samples generated in the upper layer and using them in the final estimators
 in the lower layer.</span>^{<span style="font-size:10.5pt;line-height:115%;font-family:"Arial",sans-serif;color:#1f497d"><u></u><u></u></span>}</p>
<p class="MsoNormal" align="right" style="text-align:right;line-height:115%">
^{<span style="font-size:10.5pt;line-height:115%;font-family:"Arial",sans-serif;color:#595959">Texto informado pelo autor</span>}^{<span style="font-size:10.5pt;line-height:115%;font-family:"Arial",sans-serif;color:#3b3838">.
</span>}<u></u><u></u></p>
<p class="m_-1617477441496007571xmsonormal" style="line-height:115%">
<b><span style="font-size:5.0pt;line-height:115%;color:#1f497d"><u></u> <u></u></span></b></p>
<p class="m_-1617477441496007571xmsonormal" style="line-height:115%">
<b><span style="font-size:10.5pt;line-height:115%;font-family:"Arial",sans-serif;color:#1f497d">Quando e Onde:</span></b><u></u><u></u></p>
<p class="m_-1617477441496007571xmsonormal" style="line-height:115%">
<span style="font-size:10.5pt;line-height:115%;font-family:"Arial",sans-serif;color:#595959">26 de Agosto de 2021, às
<b>16h</b></span><span style="color:#595959"><u></u><u></u></span></p>
<p class="m_-1617477441496007571xmsonormal" style="line-height:115%">
<span style="font-size:10.5pt;line-height:115%;font-family:"Arial",sans-serif;color:#595959">Via Zoom</span><span style="color:#595959"><u></u><u></u></span></p>
<p class="MsoNormal" style="line-height:115%">
<span style="font-size:10.5pt;line-height:115%;font-family:"Arial",sans-serif;color:#595959">Registre-se com antecedência para este seminário:
</span><u></u><u></u></p>
<p class="MsoNormal">
<a href="https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Ffgv-br.zoom.us%2Fmeeting%2Fregister%2FtJcsceyupjIjHtK9BmjGS2MH7CpndEK6GGZc&data=04%7C01%7Cemap%40fgv.br%7C286d9061b6eb4125e62808d96621f66f%7C79f6b639ab1242808077bdbeef869b33%7C0%7C0%7C637653118362631299%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=ZRzYvRoQVV1cgfpQ6%2BgS%2BeiHVQEEA6nJA1DsfS9Iw9Q%3D&reserved=0" target="_blank"><span style="font-size:10.0pt;font-family:"Arial",sans-serif;color:#0e71eb">https://fgv-br.zoom.us/meeting/register/tJcsceyupjIjHtK9BmjGS2MH7CpndEK6GGZc
</span></a><u></u><u></u></p>
<p class="MsoNormal" style="text-align:justify;line-height:115%">
<span style="font-size:4.0pt;line-height:115%;font-family:"Arial",sans-serif;color:#1f497d"><u></u> <u></u></span></p>
<p class="MsoNormal" style="text-align:justify;line-height:115%">
<span style="font-size:10.5pt;line-height:115%;font-family:"Arial",sans-serif;color:#595959">Após o registro, você receberá um e-mail de confirmação contendo informações sobre conexão no seminário.</span><span style="font-size:10.5pt;line-height:115%;font-family:"Arial",sans-serif"><u></u><u></u></span></p>
<p class="MsoNormal" style="text-align:justify;line-height:115%">
<span style="font-size:11.0pt;line-height:115%;font-family:"Calibri",sans-serif;color:#1f497d"><u></u> <u></u></span></p>
<p class="m_-1617477441496007571xxmsonormal1" style="line-height:115%">
<b><span style="font-size:10.5pt;line-height:115%;font-family:"Arial",sans-serif;color:#1f4e79">Palestrante:</span></b><u></u><u></u></p>
<p class="MsoNormal" style="text-align:justify;line-height:115%">
<b><span style="font-size:10.5pt;line-height:115%;font-family:"Arial",sans-serif;color:#595959">Luca Martino
</span></b><span style="font-size:10.5pt;line-height:115%;font-family:"Arial",sans-serif;color:#595959">- I have obtained the MSc degree in Electronic Engineering at the Politecnico di Milano and my PhD inStatistical Signal Processing from Universidad Carlos
 III de Madrid, Spain, in 2011. I was Assistant Professor in the Department of Signal Theory and Communications at Universidad Carlos III deMadrid since then. In August 2013 I joined the Department of Mathematics and Statistics at theUniversity of Helsinki
 with a postdoctoral position. In March 2015, I joined the Universidade de São Paulo (USP) as postdoctoral researcher. In May 2016, I started to work as a postdoctoral researcher at the University of Valencia. Currently, I am Associate Professor in URJC ("Profesor
 Titular" in Spain, permanent position).  My research interests focus mainly on Bayesian inference, Computational algorithms, Gaussian Processes and machine learning techniques.</span><span style="font-size:11.0pt;line-height:115%;font-family:"Calibri",sans-serif;color:#1f497d"><u></u><u></u></span></p>
</td>
<td width="40" style="width:30.0pt;padding:0cm 0cm 0cm 0cm"></td>
</tr>
</tbody>
</table>
</div>
<p class="MsoNormal" align="center" style="text-align:center">
<u></u><u></u></p>
</td>
<td width="0%" style="width:.06%;padding:0cm 0cm 0cm 0cm;height:27.0pt"></td>
</tr>
<tr>
<td width="100%" colspan="3" style="width:100.0%;padding:0cm 0cm 0cm 0cm">
<div align="center">
<table border="0" cellspacing="0" cellpadding="0" width="0" style="width:450.0pt;background:white">
<tbody>
<tr>
<td width="600" style="width:450.0pt;padding:0cm 0cm 0cm 0cm">
<p class="m_-1617477441496007571xmsonormal">
<span style="font-size:13.5pt;font-family:"Times New Roman",serif;color:black"><img border="0" width="599" height="66" style="width:6.2416in;height:.6916in" id="m_-1617477441496007571Imagem_x0020_2" src="cid:17b72e013b85b006a2" alt="Footer"></span><u></u><u></u></p>
</td>
</tr>
</tbody>
</table>
</div>
<p class="MsoNormal" align="center" style="text-align:center">
<span style="display:none"><u></u> <u></u></span></p>
<div align="center">
<table border="0" cellspacing="0" cellpadding="0" width="0" style="width:450.0pt">
<tbody>
<tr>
<td style="padding:0cm 0cm 0cm 0cm">
<p class="m_-1617477441496007571xmsonormal">
<span style="font-size:13.5pt;font-family:"Times New Roman",serif;color:black"><img border="0" width="599" height="1" style="width:6.2416in;height:.0083in" id="m_-1617477441496007571Imagem_x0020_3" src="cid:17b72e013b8692e333" alt="linha"></span><u></u><u></u></p>
</td>
</tr>
</tbody>
</table>
</div>
<p class="MsoNormal" align="center" style="text-align:center">
<span style="display:none"><u></u> <u></u></span></p>
<div align="center">
<table border="0" cellspacing="0" cellpadding="0" width="0" style="width:450.0pt">
<tbody>
<tr>
<td style="background:white;padding:0cm 0cm 0cm 0cm">
<p class="m_-1617477441496007571xmsonormal">
<a href="https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fportal.fgv.br%2F&data=04%7C01%7Cemap%40fgv.br%7C286d9061b6eb4125e62808d96621f66f%7C79f6b639ab1242808077bdbeef869b33%7C0%7C0%7C637653118362641292%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=TQoLDALYpT5QpN3dmUIJailzSiiz1osEtyjgUp1Hvus%3D&reserved=0" target="_blank"><span style="font-size:13.5pt;font-family:"Times New Roman",serif;color:blue;text-decoration:none"><img border="0" width="390" height="53" style="width:4.0666in;height:.55in" id="m_-1617477441496007571Imagem_x0020_4" src="cid:17b72e013b8772f6c4" alt="DICOM"></span></a><u></u><u></u></p>
</td>
<td style="background:white;padding:0cm 0cm 0cm 0cm">
<p class="m_-1617477441496007571xmsonormal">
<a href="https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.linkedin.com%2Fcompany%2Ffgv&data=04%7C01%7Cemap%40fgv.br%7C286d9061b6eb4125e62808d96621f66f%7C79f6b639ab1242808077bdbeef869b33%7C0%7C0%7C637653118362651287%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=bfTUH6FBMBn3LbvGpdfkhhd3qXlo6iwitXqC%2BZIAvuQ%3D&reserved=0" target="_blank"><span style="font-size:13.5pt;font-family:"Times New Roman",serif;color:blue;text-decoration:none"><img border="0" width="23" height="53" style="width:.2416in;height:.55in" id="m_-1617477441496007571Imagem_x0020_5" src="cid:17b72e013b88546ed5" alt="LinkedIn"></span></a><u></u><u></u></p>
</td>
<td style="background:white;padding:0cm 0cm 0cm 0cm">
<p class="m_-1617477441496007571xmsonormal">
<a href="https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.facebook.com%2FFGV%2F&data=04%7C01%7Cemap%40fgv.br%7C286d9061b6eb4125e62808d96621f66f%7C79f6b639ab1242808077bdbeef869b33%7C0%7C0%7C637653118362651287%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=diCHZhQqQ4%2F%2B6RlcQjYNEvGaVj4sEMfp6SAaNcBAi70%3D&reserved=0" target="_blank"><span style="font-size:13.5pt;font-family:"Times New Roman",serif;color:blue;text-decoration:none"><img border="0" width="23" height="53" style="width:.2416in;height:.55in" id="m_-1617477441496007571Imagem_x0020_6" src="cid:17b72e013b8935e6e6" alt="LinkedIn"></span></a><u></u><u></u></p>
</td>
<td style="background:white;padding:0cm 0cm 0cm 0cm">
<p class="m_-1617477441496007571xmsonormal">
<a href="https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.instagram.com%2FFGV.OFICIAL%2F&data=04%7C01%7Cemap%40fgv.br%7C286d9061b6eb4125e62808d96621f66f%7C79f6b639ab1242808077bdbeef869b33%7C0%7C0%7C637653118362661285%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=a4KhYzXkdSnaikwBTOFjImvU6CpZT1WhH3cqEL2wMZk%3D&reserved=0" target="_blank"><span style="font-size:13.5pt;font-family:"Times New Roman",serif;color:blue;text-decoration:none"><img border="0" width="23" height="53" style="width:.2416in;height:.55in" id="m_-1617477441496007571Imagem_x0020_7" src="cid:17b72e013b8a175ef7" alt="Facebook"></span></a><u></u><u></u></p>
</td>
<td style="background:white;padding:0cm 0cm 0cm 0cm">
<p class="m_-1617477441496007571xmsonormal">
<a href="https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Ftwitter.com%2Ffgv&data=04%7C01%7Cemap%40fgv.br%7C286d9061b6eb4125e62808d96621f66f%7C79f6b639ab1242808077bdbeef869b33%7C0%7C0%7C637653118362671280%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=rM9gjc%2BTs2dA1RSe1iARJ1MMgncKIbZSyHCyKds1q0U%3D&reserved=0" target="_blank"><span style="font-size:13.5pt;font-family:"Times New Roman",serif;color:blue;text-decoration:none"><img border="0" width="23" height="53" style="width:.2416in;height:.55in" id="m_-1617477441496007571Imagem_x0020_8" src="cid:17b72e013b8af8d708" alt="Instagram"></span></a><u></u><u></u></p>
</td>
<td style="background:white;padding:0cm 0cm 0cm 0cm">
<p class="m_-1617477441496007571xmsonormal">
<a href="https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.tiktok.com%2F%40fgv.oficial&data=04%7C01%7Cemap%40fgv.br%7C286d9061b6eb4125e62808d96621f66f%7C79f6b639ab1242808077bdbeef869b33%7C0%7C0%7C637653118362671280%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=GBYBgeDFWIPkBCbZXcRkL14dkf9mGUkuURbZ%2BB3iBNA%3D&reserved=0" target="_blank"><span style="font-size:13.5pt;font-family:"Times New Roman",serif;color:blue;text-decoration:none"><img border="0" width="23" height="52" style="width:.2416in;height:.5416in" id="m_-1617477441496007571Imagem_x0020_9" src="cid:17b72e013b8bda4f19" alt="Twitter"></span></a><u></u><u></u></p>
</td>
<td style="background:white;padding:0cm 0cm 0cm 0cm">
<p class="m_-1617477441496007571xmsonormal">
<a href="https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.youtube.com%2Ffgv&data=04%7C01%7Cemap%40fgv.br%7C286d9061b6eb4125e62808d96621f66f%7C79f6b639ab1242808077bdbeef869b33%7C0%7C0%7C637653118362681272%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=JueT2Jq9WGwDHnZwx1wis9yKOHHsZiEMjI4w4b6F8PY%3D&reserved=0" target="_blank"><span style="font-size:13.5pt;font-family:"Times New Roman",serif;color:blue;text-decoration:none"><img border="0" width="23" height="52" style="width:.2416in;height:.5416in" id="m_-1617477441496007571Imagem_x0020_10" src="cid:17b72e013b81f3aa07a" alt="YouTube"></span></a><u></u><u></u></p>
</td>
<td style="background:white;padding:0cm 0cm 0cm 0cm">
<p class="m_-1617477441496007571xmsonormal">
<a href="https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fportal.fgv.br%2Fredes-sociais&data=04%7C01%7Cemap%40fgv.br%7C286d9061b6eb4125e62808d96621f66f%7C79f6b639ab1242808077bdbeef869b33%7C0%7C0%7C637653118362691272%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=er9yWPuQZuSClhLbDbyG1M3z4Y3HkZEmkAwvB8buB14%3D&reserved=0" target="_blank"><span style="font-size:13.5pt;font-family:"Times New Roman",serif;color:blue;text-decoration:none"><img border="0" width="17" height="53" style="width:.175in;height:.55in" id="m_-1617477441496007571Imagem_x0020_11" src="cid:17b72e013b8201c188b" alt="Mais"></span></a><u></u><u></u></p>
</td>
<td style="background:white;padding:0cm 0cm 0cm 0cm">
<p class="m_-1617477441496007571xmsonormal">
<a href="https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fportal.fgv.br%2Fredes-sociais&data=04%7C01%7Cemap%40fgv.br%7C286d9061b6eb4125e62808d96621f66f%7C79f6b639ab1242808077bdbeef869b33%7C0%7C0%7C637653118362691272%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=er9yWPuQZuSClhLbDbyG1M3z4Y3HkZEmkAwvB8buB14%3D&reserved=0" target="_blank"><span style="font-size:13.5pt;font-family:"Times New Roman",serif;color:blue;text-decoration:none"><img border="0" width="50" height="53" style="width:.525in;height:.55in" id="m_-1617477441496007571Imagem_x0020_12" src="cid:17b72e013b820fd909c" alt="Mais"></span></a><u></u><u></u></p>
</td>
</tr>
</tbody>
</table>
</div>
<p class="MsoNormal" align="center" style="text-align:center"><u></u><u></u></p>
</td>
</tr>
</tbody>
</table>
</td>
<td width="0%" style="width:.04%;padding:0cm 0cm 0cm 0cm"></td>
</tr>
<tr style="height:27.0pt">
<td width="0%" style="width:.1%;padding:0cm 0cm 0cm 0cm;height:27.0pt"></td>
<td width="99%" style="width:99.86%;padding:0cm 0cm 0cm 0cm;height:27.0pt">
<p class="MsoNormal">
<u></u> <u></u></p>
<p class="MsoNormal">
<span style="font-size:10.0pt;color:#1f497d"><u></u> <u></u></span></p>
</td>
<td width="0%" style="width:.04%;padding:0cm 0cm 0cm 0cm;height:27.0pt"></td>
</tr>
<tr>
<td width="99%" colspan="2" style="width:99.96%;padding:0cm 0cm 0cm 0cm"></td>
<td width="0%" style="width:.04%;padding:0cm 0cm 0cm 0cm"></td>
</tr>
<tr>
<td width="0%" style="width:.1%;padding:0cm 0cm 0cm 0cm"></td>
<td width="99%" style="width:99.86%;padding:0cm 0cm 0cm 0cm"></td>
<td width="0%" style="width:.04%;padding:0cm 0cm 0cm 0cm"></td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<p class="MsoNormal"><u></u> <u></u></p>
</div>
</div>

</div><br clear="all"><div><br></div>-- <br><div dir="ltr" class="gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div dir="ltr" style="color:rgb(80,0,80)"><div dir="ltr">Luiz Max Fagundes de Carvalho</div><div dir="ltr">Lecturer, School of Applied Mathematics (EMAp), Getúlio Vargas Foundation (FGV), Brazil.</div><div dir="ltr">Praia de Botafogo, 190, Sala 511, Rio de Janeiro, RJ 22250-900.</div><div dir="ltr">Brazil</div></div></div></div></div></div></div>