@ale__palmaa Profile picture

Alessandro Palma

@ale__palmaa

PhD student at @HelmholtzMunich and @TU_Muenchen | ML and computational biology

Similar User
Hana Aliee photo

@AlieeHana

Lukas Heumos photo

@LukasHeumos

Leon Hetzel photo

@leon_het

Till Richter photo

@TillRichter6

Anna Schaar photo

@AnnaCSchaar

Tingying Peng photo

@p_tingying

Laura Martens photo

@lauradmartens

Vladimir Shitov photo

@shitov_happens

Luca Eyring photo

@LucaEyring

Sabrina Richter photo

@_sabrinarichter

David Fischer photo

@davidsebfischer

Dominik Klein photo

@Dominik1Klein

Giovanni Palla photo

@g_palla1

Felix Drost photo

@DrostFelix

Yuge Ji photo

@_yji_

Alessandro Palma Reposted

1/6 Looking for neural estimators of entropic #OptimalTransport or simply cool applications of #FlowMatching? Excited by novel generative modeling tools for #SingleCell data? Check out our #GENOT #NeurIPS paper tinyurl.com/yc5deeke!

Tweet Image 1

Alessandro Palma Reposted

Deep learning with differential privacy can protect sensitive information of individuals. But what about groups of multiple users? We answer this question in our #NeurIPS2024 paper arxiv.org/abs/2403.04867 Joint work w/ @mihail_sto @ArthurK48147 @guennemann #Neurips (1/7)


Alessandro Palma Reposted

Do you run functional assays? Wish you could get more results without having to scale? If you’re not using Prophet, you’re leaving potential on the table. (Warning: pitch not tweetorial)


Alessandro Palma Reposted

We introduce Spatio-Spectral GNNs (S²GNNs) – an effective modeling paradigm via the synergy of spatially and spectrally parametrized graph conv. S²GNNs generalize the spatial + FFT conv. of State Space Models like H3/Hyena. Joint work w/ @ArthurK48147 @dan1elherbst @guennemann


Alessandro Palma Reposted

Mixed Models with Multiple Instance Learning (MixMIL) received an Oral & Outstanding Student Paper award at @aistats_conf last week! 🏆 MixMIL enables accurate & interpretable patient label prediction from single-cell data by adding attention to GLMMs.#singlecell #MachineLearning

Tweet Image 1
Tweet Image 2
Tweet Image 3

Alessandro Palma Reposted

Dealing with undesired distribution shifts in unpaired translation tasks? Our #ICLR2024 paper shows how to mitigate them leveraging Unbalanced OT! We propose a method to incorporate unbalancedness into any neural Monge map estimator and demonstrate the benefits of unbalancedness.

Tweet Image 1

Alessandro Palma Reposted

Curious about how Self-Supervised Learning (SSL) is reshaping Single-Cell Genomics (SCG)? 🧬🤖 Our latest paper, "Delineating the Effective Use of Self-Supervised Learning in Single-Cell Genomics," offers an in-depth analysis. Thread [1/n] bioRxiv: doi.org/10.1101/2024.0…


Alessandro Palma Reposted

I feel delighted about "Attention-based Multi-instance Mixed Models" being accepted to @aistats_conf as an oral! It is our attention-based MIL w/ @MaxIlse on steroids: with pre-trained embeddings, variational inference and SOTA incl. histopathology! Paper: arxiv.org/abs/2311.02455


Alessandro Palma Reposted

I'm super excited to announce our new framework for exploratory electronic health record analysis "ehrapy". Although analysis is standardized for single-cell by seurat, bioconductor and scanpy, EHR analysis was until now the wild west. medrxiv.org/content/10.110…

Tweet Image 1

Alessandro Palma Reposted

Does graph contrastive learning truly improve adversarial robustness? We answer this question in our work at #GLFrontiers @ #NeurIPS2023. Paper: arxiv.org/abs/2311.17853 Poster: 16 Dec 4:30pm Hall C2 Joint work with @ZinuoYi, @AnnaStrvt, @RafikiMazen, @geisler_si, @guennemann

Tweet Image 1

Alessandro Palma Reposted

Is binarization of scATAC-seq data really necessary? The conclusion from our analysis is that a quantitative treatment is in fact beneficial. Now out in Nature Methods! @gagneurlab @fabian_theis nature.com/articles/s4159… Many additions since the preprint 👇(1/n)

Tweet Image 1

Alessandro Palma Reposted

We will present our work “The geometry of hidden representations of large transformer models” at #NeurIPS2023 Great effort from @lucrevaleriani @DiegoDoimo #fracuturello @ansuin at @AreaSciencePark, and great collaboration with Ale Laio @SISSAschool! 📝 arxiv.org/pdf/2302.00294…

Tweet Image 1

Alessandro Palma Reposted

Want to learn cell state dynamics beyond RNA velocity? Led by @PhilippWeiler7 & @MariusLange8, and with @dana_peer, our new CellRank 2 enable fate mapping leveraging pseudotime, gene scores, time points or metabolic labeling in multi-view single-cell data. biorxiv.org/content/10.110…

Tweet Image 1

Alessandro Palma Reposted

1/10 image-based screening advanced drug discovery but scaling to massive perturbation space is hard! Given a cell image, we asked if we could predict the morphological effect induced by a perturbation! Led by @ale__palmaa & w @fabian_theis we propose IMPA tinyurl.com/yy4jfn4h


Alessandro Palma Reposted

Introducing inVAE: Leveraging domain variability to learn conditionally invariant representations to unwanted biases. inVAE captures biological variations in single-cell datasets obtained from diverse conditions and labs. @HelmholtzMunich @KaplFer @soroorh1 @fabian_theis (1/7)🧵

Tweet Image 1

Alessandro Palma Reposted

🎉Wow!🎉 Our paper 'Topological Singularity Detection At Multiple Scales' was accepted at #ICML2023. 👉We question the manifold hypothesis & develop a method to detect singularities, i.e. points that violate this 'manifoldness' assumption. #topology #MachineLearning 🧵1/n

Tweet Image 1

Alessandro Palma Reposted

Fresh on bioRxiv: @sophiaMaedler and I developed ✨SPARCS✨, a technology for microscopy-based genome-wide genetic screens in human cells🔬🧬 biorxiv.org/content/10.110… This work was enabled by the amazing collaborative environment created by @hornung_lab and @labs_mann 🧵👇

Tweet Image 1

Alessandro Palma Reposted

1/10 Looking for a tool to map cells across time and space? We introduce moscot-tools.org, a scalable framework for #optimaltransport (#OT) applications in single-cell genomics! bit.ly/3MiICFc

Tweet Image 1

Loading...

Something went wrong.


Something went wrong.