Marco Fraccaro

30 posts

Marco Fraccaro

@fraccarom

Machine learning researcher

Katılım Haziran 2016

152 Takip Edilen390 Takipçiler

Sabitlenmiş Tweet

Marco Fraccaro@fraccarom·2 Ara

My PhD thesis on "Deep Latent Variable Models for Sequential Data" is finally online! If you want to learn how VAEs can be extended to the sequential setting you should check it out: marcofraccaro.github.io/download/publi… Special thanks to my amazing supervisors @OleWinther1 @ulrichpaquet

English

133

596

Marco Fraccaro@fraccarom·26 Mar

Really cool to see deep sequential latent variable models used to learn to fly real drones - not in a simulated environment!!

Justin Bayer@usuallyuseless

Variational state-space models can be used to perform optimal control of drones only equipped with LiDARs. So show my dear colleagues Philip Becker-Ehmck and Max Karl in their latest work "Learning to fly". arxiv.org/abs/2003.08876. Thread.

English

Marco Fraccaro@fraccarom·11 Ara

You can also try BIVA yourself, with our Tensorflow AND PyTorch implementations! Tensorflow code: github.com/larsmaaloee/BI… PyTorch code: github.com/vlievin/biva-p…

English

Marco Fraccaro@fraccarom·11 Ara

VAEs enthusiasts, if you are at #NeurIPS2019 you should come to poster 110 today at 10.45. We'll be presenting BIVA, and show you how to build powerful deep hierarchical VAEs. Paper: papers.nips.cc/paper/8882-biv… @larsmaaloe @valentinlievin @OleWinther1

English

Marco Fraccaro@fraccarom·13 Ağu

It was a pleasure to talk about deep latent variable models at the summer school on generative models by @DTU_Compute and @uni_copenhagen! If none among the 1000s of VAE tutorials out there is good enough for you, try my slides summer-school-gan.compute.dtu.dk/slides_marco.p…

English

Marco Fraccaro retweetledi

Lars Kai Hansen@Tweetteresearch·5 Ağu

Be part of the expanding AI hub in Copenhagen - 12 Postdoc positions as part of multidisciplinary university collaboration. DM if you are interested dtu.dk/english/about/… @DTUtweet @DTU_Compute @jan_madsen

English

Marco Fraccaro@fraccarom·31 Tem

Great to see this working with multiple objects!

Silvia Chiappa@csilviavr

New paper with @ulrichpaquet using exact temporal inference for learning the dynamics of moving objects from pixels.

English

Marco Fraccaro retweetledi

Yarin@yaringal·14 Haz

Really excited to release Bayesian Deep Learning Benchmarks - please share with others who you think might like this, and have a look at the blog/repo/colab: github.com/OATML/bdl-benc… This work was done over a period of a year and a half by many collaborators @OATML_Oxford

English

378

1.1K

Marco Fraccaro retweetledi

Martina Fischetti@MartiFischetti·10 Şub

My PhD in a minute 😊 Thanks @Vattenfall_DK #phd1minute #orms #appliedOR #operationsresearch #GreenEnergy

English

Jakub Tomczak@jmtomczak·9 Şub

This paper: arxiv.org/abs/1902.02102 (by @larsmaaloe @fraccarom @LieValentin & @OleWinther1) was put on arXiv two days ago and it was already discussed at @UvA_Amsterdam. It is wonderful and extremely well-written, all experiments are carefully carried out. Highly recommended!

English

Marco Fraccaro@fraccarom·9 Şub

@jmtomczak @larsmaaloe @LieValentin @OleWinther1 @UvA_Amsterdam Thanks! Glad you liked it

English

Marco Fraccaro@fraccarom·7 Şub

"BIVA: A Very Deep Hierarchy of Latent Variables for Generative Modeling". A skip-connected generative model and a bidirectional inference network are all you need! Work with @larsmaaloe @LieValentin @OleWinther1 arxiv.org/abs/1902.02102

English

Marco Fraccaro retweetledi

Papers with Code@paperswithcode·1 Şub

We’ve just released the new Papers With Code! Site now has over 950+ ML tasks, 500+ evaluation tables (including state of the art results) and 8500+ papers with code. Explore the resource here: paperswithcode.com/sota. Have fun!

English

752

1.5K

Marco Fraccaro retweetledi

Martina Fischetti@MartiFischetti·4 Oca

Excited to announce that the project I am leading in #Vattenfall is among the 2019 @INFORMS Franz Edelman Award finalists!Thanks to #orms, @VattenfallBXL saved $170 million in offshore wind, thus contributing to a fossil free future.With Kristoffersen,Hjort,Pisinger,@mmonaci74

INFORMS@INFORMS

INFORMS is happy to announce the six 2019 Franz Edelman Award Finalists! Congratulations to the finalists for having outstanding #ORMS & #Analytics projects. bit.ly/2CNyfFy

English

Marco Fraccaro retweetledi

Ulrich Paquet@ulrichpaquet·19 Ara

Making probabilistic deep generative models... compositional! With Sumedh K. Ghaisas and Olivier Tieleman @DeepMindAI. Finally out on arXiv today 😀. arxiv.org/abs/1812.07480

English

143

Marco Fraccaro retweetledi

Ulrich Paquet@ulrichpaquet·30 Eki

For the technically-minded: Riemannian Manifold Hamiltonian Monte Carlo for Gaussian Process Models arxiv.org/abs/1810.11893 An (old) tech report with my good friend @fraccarom, finally seeing the arXiv light of day!! 😀

English

Marco Fraccaro retweetledi

Matteo Fischetti@MFischetti·13 Eyl

.@MartiFischetti @fraccarom @mmonaci74 and David Pisinger won the 2018 AIRO Best Application-Oriented Paper award for their work on wind-farm optimization #ods2018 #orms

English

Marco Fraccaro retweetledi

Ali Eslami@arkitus·11 Tem

Come see our poster on Generative Temporal Models at @icmlconf: Hall B poster 101. @fraccarom woop woop!

English

Marco Fraccaro@fraccarom·15 Haz

Great work by @arkitus @DeepSpiker @fabiointheuk! GQNs are also an important component of our latest ICML paper on generative temporal models with spatial memory! x.com/arkitus/status…

Ali Eslami@arkitus

"Neural scene representation and rendering" now in @sciencemagazine. By training deep networks to predict what scenes look like from new viewpoints, we get them to understand images: deepmind.com/blog/neural-sc… @DeepSpiker @OriolVinyalsML @theophaneweber @demishassabis

English

Marco Fraccaro retweetledi

Danilo J. Rezende@DaniloJRezende·7 Haz

The paper Generative Temporal Models with Spatial Memory for Partially Observed Environments arxiv.org/abs/1804.09401 was accepted as a long talk at @icmlconf Congrats again @fraccarom ! x.com/DeepSpiker/sta…

Danilo J. Rezende@DaniloJRezende

Generative Temporal Models with Spatial Memory for Partially Observed Environments arxiv.org/abs/1804.09401 It was really cool working with Marco Fraccaro on this project during his internship @DeepMindAI. Great work Marco! #statespacemodels #inference #spatialmemory #vae

English

Keşfet

@larsmaaloe @valentinlievin @OleWinther1 @DTU_Compute @uni_copenhagen @DTUtweet @jan_madsen @OATML_Oxford