Charlotte Bunne
@_bunnech
Followers
3,436
Following
489
Media
43
Statuses
481
PostDoc at @Genentech and @Stanford and Incoming Assistant Professor at @EPFL in Computer Science and Life Sciences.
🇨🇭and 🇺🇸
Joined November 2015
Don't wanna be here?
Send us removal request.
Explore trending content on Musk Viewer
Renzo
• 149514 Tweets
#WWERaw
• 82544 Tweets
#GodMorningTuesday
• 37096 Tweets
Scarlett Johansson
• 33379 Tweets
Oilers
• 32756 Tweets
Fani Willis
• 31365 Tweets
定額減税
• 30436 Tweets
Kingdom Hearts
• 28122 Tweets
राजीव गांधी
• 26553 Tweets
梅雨入り
• 25308 Tweets
Birds Nurturing
• 24042 Tweets
給与明細
• 24010 Tweets
Gunther
• 22163 Tweets
#Canucks
• 20882 Tweets
Vancouver
• 20788 Tweets
#ファンパレハーフアニバーサリー
• 18934 Tweets
Edmonton
• 17581 Tweets
ドジャース
• 17426 Tweets
金額明記
• 17231 Tweets
Amber Rose
• 15464 Tweets
ショートアニメ
• 14937 Tweets
キングダムハーツ
• 14089 Tweets
Lyra
• 12452 Tweets
国民実感
• 10659 Tweets
Pinned Tweet
Super excited to join
@EPFL_en
as assistant professor in the fall of 2024. Heartfelt gratitude to everyone who supported me!
I am
#hiring
#PhDs
and
#PostDocs
! Interested in developing machine learning methods for enhancing science🧬and personalized medicine💊?
Details in the🧵!
📢 The
@ETH_Rat
has announced the appointment of two professors at EPFL. Congratulations to Charlotte Bunne
@_bunnech
and Wenzel Jakob
@wenzeljakob
.
0
4
53
53
69
755
How does optimal transport allow us to study dynamical systems? How does it connect to control theory, flow matching, & diffusion models? How is it advancing molecular biology research?
Find the answers in our tutorial! Recording, slides, & script under
.
5
137
738
🎉I defended my
#PhD
!
🧵Neural optimal transport for biomedicine? A thread below with the
#thesis
highlights!
But first, a huge thank you to my advisors
@arkrause
and
@cuturimarco
, my committee
@lucaspelkmans
and
@jure
, and all collaborators for such an incredible adventure!
21
42
534
✨Another edition of the Optimal Transport and Machine Learning workshop is taking place at
#NeurIPS2023
.
🚀Contribute your work on OT and theory, computation, deep learning, or applications until Sept 29 ()! Details on speakers and schedule coming soon.
2
35
221
Excited about neural
#optimaltransport
methods for modeling
#singlecell
perturbation responses but wondering how to integrate cell birth and death?
@frederikelubeck
has the answer for you in the
#AI4science
workshop
@NeurIPSConf
! Join us in rooms 388 - 390.
2
27
184
Join me at my poster
#1035
in Hall J now to hear about conditional neural
#optimaltransport
with applications in single-cell biology.
3
10
133
Check out our work on proximal optimal transport modeling of population dynamics (JKOnet) at the
#TimeSeriesWorkshop
at
#ICML2021
.
Joint work with
@laetitia_ppx
,
@arkrause
, and
@cuturimarco
. 🧵
Paper:
4
36
127
Join us on October 21, 2022, at
@MIT
for the first iteration of the Molecular ML Conference (MoML) held in memory of
@OctavianEGanea
. Registration and poster submissions have started already!
0
25
118
The Optimal Transport and Machine Learning
#OTML
workshop
@NeurIPSConf
#NeurIPS2021
is now open for submissions via
@OpenReviewnet
. Deadline is September 18, 2021!
We look forward to your contributions!
0
36
119
Check out our work on predicting
#singlecell
perturbation responses using neural optimal transport (
#CellOT
), which we presented at the
#OTML
Workshop at
@NeurIPS
2021.
Paper:
1
23
111
Done with
#ICLR2024
and almost ready for
#AISTATS2024
? Then consider submitting to two fantastic workshops at
#NeurIPS2023
! Deadline October 3.
⌛️ Workshop on Optimal Transport and Machine Learning:
🍃 Workshop on Diffusion Models:
0
15
91
Join us at our tutorial at
#ICML2023
! Today, 9:30 am local time! 🚀
We will cover various flavors of
#optimaltransport
, dive into neural OT solvers, and discuss dynamic OT formulations, incl. links to neural ODEs, SDEs, PDEs, flow matching, and diffusion models!
3
10
84
I will be giving an Invited Talk tomorrow at
@appliedmldays
about my recent work on 'Optimal Transport Modeling of Population Dynamics' with applications in
#singlecell
biology
#AMLDEPFL22
.
2
11
82
Honored to announce that we received the best paper award! Stay tuned for updates on our work.
Thanks again to my co-authors
@laetitia_ppx
,
@arkrause
, and
@cuturimarco
and to the organizers (
@yuqirose
,
@lovvge
) for assembling this year’s
#TimeSeriesWorkshop
#ICML2021
#TSW2021
.
Check out our work on proximal optimal transport modeling of population dynamics (JKOnet) at the
#TimeSeriesWorkshop
at
#ICML2021
.
Joint work with
@laetitia_ppx
,
@arkrause
, and
@cuturimarco
. 🧵
Paper:
4
36
127
2
7
80
✨Call for Papers✨
We are now accepting submissions for our
#Frontiers4LCD
workshop at
#ICML2023
!
Guidelines for the authors can be found here
Submission deadline: May 29, 2023
2
11
76
Molecules are largely unaltered during chemical reactions. In our new work () on graph-based retrosynthesis prediction, we identify reusable substructures within molecules to produce reactants. Led by
@_vsomnath
(w/
@cwcoley
,
@arkrause
,
@barzilayregina
).
0
14
63
Join the Optimal Transport and Machine Learning
#OTML2021
workshop at
@NeurIPSConf
2021 starting now!
2
9
57
We are honoured to receive the best paper award! Thanks for organising the workshop.
#NeurIPS2018
#R2L
2
5
52
Join us for the 1st Machine Learning for Genomics Explorations (
#MLGenX
) workshop at
#ICLR2024
on May 11, 2024 in Vienna 🇦🇹!
✨Stay tuned for fantastic keynotes on foundation models, generalizability, and causal methods for genomics!
🗒️Working in this space? Then submit a paper!
📢🚨 Excited to announce the Machine Learning for Genomics Explorations (MLGenX) workshop at
@iclr_conf
2024.
⏰ Submission deadline: February 4, 2024 (AOE).
🔗 Call for papers:
🗓️ Date: May 11, 2024.
1
9
31
1
10
50
Happening today!
Measuring how drugs affect cells often means destroying them, making it difficult to measure single-cell dynamics. Next week at M2D2,
@_bunnech
will explain how Optimal Transport can help re-align unpaired snapshots from different cells. Join us Tues (June 14) at 10am ET!
1
13
52
0
4
38
Join us today at our poster at
#AISTATS2022
at 1:30 pm CEST or 7:30 am EST.
Now with exciting results on robustness to noise and
#singlecell
#genomics
!
Check out our work on proximal optimal transport modeling of population dynamics (JKOnet) at the
#TimeSeriesWorkshop
at
#ICML2021
.
Joint work with
@laetitia_ppx
,
@arkrause
, and
@cuturimarco
. 🧵
Paper:
4
36
127
1
9
31
Next is
@vdbergrianne
on “Two-for-one: Diffusion Models and Force Fields for Coarse-Grained Molecular Dynamics” in Ballroom B!
Our
#Frontiers4LCD
workshop at
#ICML
will take place on Friday from 9am-5pm, in Ballroom B!
We have an exciting line of speakers from learning, optimal control/transport, RL, Neural PDE, waiting for you 🤯😉... so drop by!
🗓️ Schedule:
0
6
41
0
2
27
Starting in 30 minutes. Join the AI in the Molecular World workshop at
@appliedmldays
#AMLDEPFL22
!
I will be giving an Invited Talk tomorrow at
@appliedmldays
about my recent work on 'Optimal Transport Modeling of Population Dynamics' with applications in
#singlecell
biology
#AMLDEPFL22
.
2
11
82
1
3
22
Introduction to
@Minio
, an S3-compatible object-storage server with first class
@Docker
support from
@alexellisuk
.
#Tutorial
0
6
19
Submissions are due in a week! 🚀 We accept long (8 pages) and short (4 pages) papers on topics related to learning, control, and dynamical systems!
See our ‘Call for Papers’ for more information:
✨Call for Papers✨
We are now accepting submissions for our
#Frontiers4LCD
workshop at
#ICML2023
!
Guidelines for the authors can be found here
Submission deadline: May 29, 2023
2
11
76
0
5
20
Very happy to be part of this effort! I hope this review will be a valuable resource for many!
Our
@NCCR_Catalysis
collaboration (with
@acvaucher
,
@ruben_laplaza
,
@_bunnech
,
@arkrause
,
@corminboeuf_lab
,
@teodorolaino
) made the cover of
@WIREs_Reviews
Computational Molecular Science!
#compchem
#AI4Science
Check it out:
3
16
73
0
1
20
We are looking for students with background in computer science and computational biology. No exposure to biology yet? Don't hesitate to apply!
Questions? Please reach out or otherwise directly apply via . The deadline is on🗓️December 15 (⏳in one week)!
2
1
19
1
6
17
How to predict
#singlecell
perturbation responses? What is optimal transport and why can it solve such puzzles? How to integrate this into
#machinelearning
pipelines to predict treatment outcomes of unseen patients? Join us tomorrow in-person or on Zoom!
📢 Nov 16
⏰ 9AM EST
💬 Oana Ursu
@oana__ursu
💡 Primer: Analytical challenges and opportunities for studying cell state transitions at single cell level
⏰ 10AM EST
💬 Charlotte Bunne
@_bunnech
💡 Neural Optimal Transport for Inferring Single-Cell Responses to Perturbations
0
2
9
1
6
16
Flood of information and inspiration in our lightning talks!
#WiDSZurich2018
#WiDS2018
© Zsofi Lang
0
4
13
With
@ValentinDeBort1
,
@guanhorng_liu
,
@iamct_r
,
@AnimaAnandkumar
,
@mraginsky
,
@pratikac
, and Melanie Zeilinger.
0
0
12
The full thesis is here: .
🚀 And of course, stay tuned for more research and updates on what's next!
0
0
11
This thesis develops
#machinelearning
methods for modeling dynamical systems. Central application:
#drugdiscovery
&🧬
#singlecell
biology with the aim to predict single cell responses to perturbations such as 💊drugs. Challenge: We cannot trace the response of individual cells.
1
1
10
I am very much looking forward to be joining such an exceptional environment and to work with many new colleagues!
Aaand, before moving to
@EPFL_en
, you can find me soon at
@genentech
and
@Stanford
, spending exciting months with Aviv Regev and
@Jure
Leskovec.
1
1
10
@LeoChan213
@arkrause
@CuturiMarco
@lucaspelkmans
@jure
Unfortunately not. But you can find a talk from last year here:
1
1
8
To achieve this, this thesis connects a variety of seemingly unrelated concepts into a unified framework: We propose
#neural
#optimaltransport
methods that can be parameterized through static maps or dynamic ODEs, SDEs, PDEs. See also for a deep dive!
1
0
9
The first method,
#CellOT
(
@naturemethods
) uses optimal transport to accurately predict single-cell responses to various drugs or developmental signals. We learn a neural
#Monge
map that transforms the unperturbed cell population into its perturbed state.
1
1
8
"Neural Relational Inference for Interacting Systems" is accepted to
@icmlconf
(w/
@EthanFetaya
, Jackson Wang,
@wellingmax
, Rich Zemel)! Graph neural nets can learn to infer latent relational structure and model multi-agent and physical dynamics 👉
3
93
338
0
1
9
Wie wähle ich in der
#Bundestagswahl
am 26. September, wenn ich im
#Ausland
lebe? Informationen vom
@Wahlleiter_Bund
!
0
1
7
0
1
8
Work done by
@octavianeganea
, Xinyuan Huang,
@yataobian
,
@barzilayregina
, Tommi Jaakkola,
@arkrause
, and myself.
1
0
8
And for more insights, have a look at
@octavianeganea
's previous 🧵!
How well can we predict protein complexes from the unbound structures in an end-to-end fashion? We design EquiDock, a deep learning model incorporating a key geometric constraint: the predicted complex is independent of the initial positions and orientations of the constituents.
5
46
232
0
0
7
With fantastic co-organizers
@elmelis
,
@korba_anna
,
@cuturimarco
, Aram-Alexandre Pooladian, and Ziv Goldfeld!
1
0
7
#CondOT
(
@neuripsconf
2022) generalizes
#CellOT
and allows conditioning the inferred responses on a context, e.g., patient meta-information. This is crucial when moving to prediction tasks on patient cohorts 🏥!
1
0
5
1
0
6
Stay tuned for updates on other data modalities, patient studies, and incorporated experimental setups, which will follow soon!
This was a great collaborative effort and joint work with
@stefangstark
*,
@gabrigut
*, ...,
@lehmannlab
,
@lucaspelkmans
,
@arkrause
, and
@gxr
.
0
0
6
Moving beyond such static formulations, we explore
#OT
's connections to dynamical systems theory.
The result is
#JKOnet
(
@aistats_conf
2022) that learns the underlying epigenetic landscape in order to predict how cells differentiate into various lineages.
1
0
5
And we all wonder why ..
Several analyses suggest that women’s publishing rate has fallen relative to men’s amid the pandemic.
2
112
195
0
0
5
Webpage for the tutorial on ‘Adversarial Robustness’ by
@aleks_madry
and
@zicokolter
#Tutorial
#NeurIPS2018
0
2
5
In
#singlecell
genomics, we observe static snapshots describing a population’s evolution on aggregate without providing insights on individual cell trajectories.
To understand and predict these complex dynamics, we need to explain the mechanism driving the population’s evolution.
1
1
4
Listen to our contributed talk at the
#TimeSeriesWorkshop
, see our paper for the details behind all of this, and get in touch if you are interested in further discussions!
0
0
4
Theoretical insights, algorithms and applications in machine learning: Computational Optimal Transport book by
@gabrielpeyre
and
@marcocuturi
.
#OptimalTransport
#MachineLearning
#Book
0
0
4
“Teaching a Variational Autoencoder (VAE) to draw MNIST characters” by Felix Mohr
0
0
3
To achieve this, we first propose a novel numerical scheme for solving JKO flows using ICNNs (by
@brandondamos
,
@chinwei_h
). We then differentiate through this reformulated JKO flow to learn the underlying energy functional explaining the dynamics.
1
0
4
Congrats, Lisa! 🎉
If you are a student passionate about computational biomedicine, don’t miss this chance and join her lab!
I am very happy to share that I am joining the Department of Infectious Diseases & Int Care
@ChariteBerlin
as a professor of Systems Biology of Infectious Diseases.
My research will focus on computational immunology & I will soon be looking for students to join me! 🙃 (1/3)
20
11
123
1
0
4
: a (quite) new journal dedicated to interactive, visual and clear explanations of topics related to machine learning research.
#MachineLearning
#SciComm
Using Artificial Intelligence to Augment Human Intelligence -- A new Distill commentary article by
@shancarter
&
@michael_nielsen
.
1
109
232
0
1
4
For those migrating to or another Mastodon server, you can also find me here: !
0
0
4
Excited to be among such a great line-up of speakers including
@mmbronstein
and Anne-Florence Bitbol giving the other two Invited Talks!
0
0
4
"Intelligente Infrastrukturen erfordern Bildung in Informatik, Technik und Humanwissenschaften" - Klaus Mainzer über KI bei
#DatenDingeWerte
0
0
3
Happening now!
How to predict
#singlecell
perturbation responses? What is optimal transport and why can it solve such puzzles? How to integrate this into
#machinelearning
pipelines to predict treatment outcomes of unseen patients? Join us tomorrow in-person or on Zoom!
1
6
16
0
0
3
How do we communicate risk and model uncertainty to patients?
@DaniCMBelg
on challenges of personalized medicine.
#PersonalizedMedicine
#DataScience
#WiDSZurich2018
0
0
3
@pariset_matteo
is at
#UAI2023
presenting work on Schrödinger bridges for (partially) aligned data and their connections to flow matching! With applications on
#singlecell
problems and
#protein
conformation change prediction!
Join him at his poster!
I'm happy to present our extension of Schrödinger Bridges to paired data at UAI 2023 in Pittsburgh. Attend our spotlight presentation today or come to the poster session to learn more!
@_vsomnath
@yapinghsieh
@_bunnech
#MachineLearning
#Diffusion
#SchrodingerBridges
0
8
62
0
0
3
Joint work with
@frederikelubeck
,
@gabrigut
, Jacobo Sarabia del Castillo,
@lucaspelkmans
,
@elmelis
, and myself.
Paper:
1
1
3
Little known fact: Brushing teeth is actually a 1D convolution. A kernel of bristles is slid across a 2x16 dental matrix.
#deeplearning
1
31
70
0
0
3
Why biology needs computer science: "Local neighborhoods surrounding 205 resting, nonmitotic cells were annotated by hand using Microsoft Powerpoint."
#Bioinformatics
0
0
3
Save the date:
#MITMath
#SimonsLecture
Series with Stephan Boyd and Sylvia Serfaty April 18-25
@MITMath
0
3
4
0
0
3
Costis Daskalakis talks about Improving GANs using Game Theory and Statistics today
@MIT_CSAIL
.
0
1
3
"Towards precision medicine with big data and molecular insight": Inaugural lecture of
@gxr
@ETH
#BiomedicalInformatics
.
0
0
2
0
0
3
Without knowing the old password, I changed the single-user password of my Macbook in a 5 minute break. I need an update..
#AppleSecurity
0
0
3
@keiththeforest
In the last years (e.g., ) all workshop streams were available after the conference. We will check with the conference organizers and leave a note on our webpage.
1
0
3
„[Schrift] entbindet uns von den Zwängen des Hier und Jetzt.” — Florian Coulmas
0
0
3
And for me: Let's get back to work..
0
0
2
0
0
2
