BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Scaling species distribution models with deep learning
DTSTART:20230706T140000
DTEND:20230706T160000
DTSTAMP:20260407T183820Z
UID:dc754e138245c2822fb7e6f7a4fc381463c5db1672ead8024d6f6700
CATEGORIES:Conferences - Seminars
DESCRIPTION:Robin Zbinden\nEDIC candidacy exam\nExam president: Prof. Mari
 a Brbic\nThesis advisor: Prof. Devis Tuia\nCo-examiner: Prof. Stéphane Jo
 ost\n\nAbstract\nSpecies distribution models (SDMs) relate environmental c
 onditions to the presence of species and play an important role in conserv
 ation ecology. Nonetheless\, the development of reliable SDMs encounters o
 bstacles such as limited data availability and selection bias\, which hind
 er their predictive capability and transferability. Recently\, advanced ma
 chine learning and deep learning approaches have emerged to address simila
 r challenges\, presenting exciting prospects to enhance the performance an
 d generalizability of SDMs. This proposal aims to investigate the viabilit
 y of employing these methods in SDMs and explore their adaptability within
  this particular domain.\nSpecifically\, we discuss three existing works a
 nd their relevance to our research. We highlight first the inherent diffic
 ulties in SDMs and explore standard approaches. We then delve into recent 
 machine learning techniques that can be leveraged to address these challen
 ges effectively. Finally\, we outline our ongoing efforts and future resea
 rch directions that aim to integrate deep learning methods into SDMs.\n\nB
 ackground papers\n\n	A maximum entropy approach to species distribution mo
 deling By Steven J. Phillips\, Miroslav Dudík\, Robert E. Schapire\n	http
 s://dl.acm.org/doi/10.1145/1015330.1015412\n	Presence-Only Geographical Pr
 iors for Fine-Grained Image Classification\, By Oisin Mac Aodha\, Elijah C
 ole\, Pietro Perona\n	https://ieeexplore.ieee.org/document/9008116\n	A Sim
 ple Framework for Contrastive Learning of Visual Representations\, By Ting
  Chen\, Simon Kornblith\, Mohammad Norouzi\, Geoffrey Hinton\n	https://
 proceedings.mlr.press/v119/chen20j.html\n
LOCATION:INM 11 https://plan.epfl.ch/?room==INM%2011
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR