Filipe Rodrigues

Associate Professor at Denmark Technical University (DTU)

[DTU Machine Learning for Smart Mobility @mlsm_dtu]

Congrats to @DanieleGammelli whom successfully defended his #PhD thesis "Learning and Control for Adaptive Transportation Systems". You will be missed! Check below for more info including the abstract and the list of 6 papers published by Daniele!
https://t.co/MxFc1HeKdb

[DTU Machine Learning for Smart Mobility @mlsm_dtu]

#JobAlert Our @mlsm_dtu group is recruiting 3 #PhDposition in exciting topics!! Take a look and apply !
1) https://t.co/KMKoLtpPH4
2) https://t.co/g72WGwTONt
3) https://t.co/CmoYdccd1l
#MachineLearning #publictransport #simulation #behaviour #traffic #autonomousdriving

[DTU Machine Learning for Smart Mobility @mlsm_dtu]

The 7th International Choice Modelling Conference (ICMC) will be held in Reykjavik, Iceland from 23-25 May 2022. Our group will contribute with six interesting papers. Check out below for more details.
https://t.co/g4ITXykNCQ #MachineLearning #ICMC #ChoiceModelling #Transport

[Daniele Gammelli @DanieleGammelli]

Predictions from ML models are often used for downstream decision-making tasks, where a sequential predict-then-optimize strategy is used to train the ML model.

But do accurate predictions translate into better decisions? Spoiler: NO

arXiv: https://t.co/ipZc7gIy4E

🧵(1/5)

[Filipe Rodrigues @fmpr87]

Is a complex deep neural network better than a simple linear regression for spatio-temporal forecasting in transportation problems? Check here how a simple linear regression can provide comparable results to many state-of-the-art approaches: https://t.co/yK0xkr10xn

[Filipe Rodrigues @fmpr87]

My paper on "Scaling Bayesian inference of mixed multinomial logit models to large datasets" is now published online: https://t.co/yVh3ogicb7
Make sure to check out the provided PyTorch implementation with an easy-to-use interface: https://t.co/b6fH855CzM

[Daniele Gammelli @DanieleGammelli]

Join us at #INNF Workshop #ICML2021 at 11am UTC (9am CET) today!

The efficiency of mobility systems depends on the ability to model the need for transportation.

We show how LVMs and normalizing flows enable us to generate distributions matching complex urban topologies.

[Daniele Gammelli @DanieleGammelli]

How can we learn *transferrable* control policies for Autonomous Mobility-on-Demand systems across cities? Our recent work shows how graph neural networks and deep RL agents exhibit promising zero-shot transfer capabilities. Up on arXiv! https://t.co/6AO6FcnkiO

[Filipe Rodrigues @fmpr87]

Cool work from our PhD student @DanieleGammelli on graph neural network reinforcement learning for autonomous mobility-on-demand systems in collaboration with @StanfordASL!

[Daniele Gammelli @DanieleGammelli]

How can we learn *transferrable* control policies for Autonomous Mobility-on-Demand systems across cities? Our recent work shows how graph neural networks and deep RL agents exhibit promising zero-shot transfer capabilities. Up on arXiv! https://t.co/6AO6FcnkiO

[Filipe Rodrigues @fmpr87]

New PhD student position open at the Machine Learning for Smart Mobility group (https://t.co/peK6rVjt3W) from the Technical University of Denmark: https://t.co/NgU66NTHTU

[DTU Machine Learning for Smart Mobility @mlsm_dtu]

Our team is ready to start the 2020 season🏃‍♀️🏃‍♂️

[DTU Machine Learning for Smart Mobility @mlsm_dtu]

Our members @inonpeled & @fmpr87 just presented the latest research on reinforcement learning for non-recurrent traffic control & uncertainty estimation in demand prediction for autonomous mobility at #IEEE #itsc2019 NZ https://t.co/wDOQqXThq9. Great growing community @itssieee

[DTU @DTUtweet]

"The paradigm shift that we believe is ideal for us as a society is where the transport system primarily adapts to us," says prof Francisco Pereira @dtumanagement. Meet him at DTU HighTech Summit. Sign up here👉 https://t.co/jyOeLsBFaN
#dkforsk #dktrp #dkdigi #HighTechDTU

[DTU Machine Learning for Smart Mobility @mlsm_dtu]

Quite productive visit to #kobecity with 6 papers and a session with the latest developments on machine learning and discrete choice modelling at #ICMC #japan https://t.co/EFAdcqkRjR

[DTU Machine Learning for Smart Mobility @mlsm_dtu]

New PhD student position open at the Machine Learning for Smart Mobility group @mlsm_dtu from the Technical University of Denmark on "Machine Learning for shared, connected and cooperative automated vehicles"
https://t.co/yNU1oo09Ch

[DTU Machine Learning for Smart Mobility @mlsm_dtu]

One more step towards future traffic control: check our latest paper on the impact of demand surges, incidents and sensor failures on deep reinforcement learning-based traffic signal control https://t.co/aF1Ie2in7a #MachineLearning #trafficcontrol #arXiv

[DTU Machine Learning for Smart Mobility @mlsm_dtu]

Congratulations to @MarkouIoulia for the her PhD dissertation!!! #congrats #throwyourhats #MLSM @dtumanagement

[DTU Machine Learning for Smart Mobility @mlsm_dtu]

Check out the 11 contributions involving @mlsm_dtu members at TRB2019! #TRb2019 #TRB19 #TRBAM
https://t.co/LidzdwEDII

[DTU Machine Learning for Smart Mobility @mlsm_dtu]

Our group's website is now online! Make sure to check it!
https://t.co/8bSBlBC5oI

[Filipe Rodrigues @fmpr87]

Check out our new AAAI 2018 paper: "Deep Learning from Crowds" (https://t.co/9wxAefrRyI). Something to consider the next time you train a deep NN on crowdsourced labels...