AIConfPaper
← Search

Andrea Dittadi

19 accepted papers

2026

Are Object-Centric Representations Better At Compositional Generalization?

Ferdinand Kapl, Amir Mohammad Karimi Mamaghan, Maximilian Seitzer, Karl Johansson, Carsten Marr, Stefan Bauer +1
ICML 2026poster

Compositional generalization, the ability to reason about novel combinations of familiar concepts, is fundamental to human cognition and a critical challenge for machine learning. Object-centric (OC) representations, which encode a scene as a set of objects, are often argued to support such generali…

Cited by 0Source ↗Scholar ↗
2026

Flow-Based Density Ratio Estimation for Intractable Distributions with Applications in Genomics

Egor Antipov, Alessandro Palma, Lorenzo Consoli, Stephan Günnemann, Andrea Dittadi, Fabian Theis
ICML 2026poster

Estimating density ratios between pairs of intractable data distributions is a core problem in probabilistic modeling, enabling principled comparisons of sample likelihoods under different data-generating processes across conditions and covariates. While exact-likelihood models such as normalizing f…

Cited by 0Source ↗Scholar ↗
2026

Logit Distance Bounds Representational Similarity

Beatrix M. G. Nielsen, Emanuele Marconato, Luigi Gresele, Andrea Dittadi, Simon Buchholz
ICML 2026poster

For a broad family of discriminative models that includes autoregressive language models, identifiability results imply that if two models induce the same conditional distributions, then their internal representations agree up to an invertible linear transformation. We ask whether an analogous concl…

Cited by 0Source ↗Scholar ↗
2026

Scalable and Interpretable Representation Alignment with Ordinal Similarity

Diogo Soares, Pankhil Gawade, Andrea Dittadi, Ewa Szczurek
ICML 2026poster

Evaluating representation similarity is fundamental to representation learning. However, existing metrics suffer from significant limitations: they are difficult to interpret due to shifting baselines, lack robustness to outliers, and are frequently computationally intractable for large datasets, fo…

Cited by 0Source ↗Scholar ↗
2025

Exploring the Effectiveness of Object-Centric Representations in Visual Question Answering: Comparative Insights with Foundation Models

Amir Mohammad Karimi Mamaghan, Samuele Papa, Karl Henrik Johansson, Stefan Bauer, Andrea Dittadi
ICLR 2025poster

Object-centric (OC) representations, which model visual scenes as compositions of discrete objects, have the potential to be used in various downstream tasks to achieve systematic compositional generalization and facilitate reasoning. However, these claims have yet to be thoroughly validated empiric…

Cited by 8Source ↗PDF ↗Scholar ↗
2025

Multi-Modal and Multi-Attribute Generation of Single Cells with CFGen

Alessandro Palma, Till Richter, Hanyi Zhang, Manuel Lubetzki, Alexander Tong, Andrea Dittadi +1
ICLR 2025poster

Generative modeling of single-cell RNA-seq data is crucial for tasks like trajectory inference, batch effect removal, and simulation of realistic cellular data. However, recent deep generative models simulating synthetic single cells from noise operate on pre-processed continuous gene expression app…

Cited by 0Source ↗PDF ↗code ↗Scholar ↗
2025

When Does Closeness in Distribution Imply Representational Similarity? An Identifiability Perspective

Beatrix Miranda Ginn Nielsen, Emanuele Marconato, Andrea Dittadi, Luigi Gresele
NeurIPS 2025poster

When and why representations learned by different deep neural networks are similar is an active research topic. We choose to address these questions from the perspective of identifiability theory, which suggests that a measure of representational similarity should be invariant to transformation…

Cited by 0Source ↗Scholar ↗
2024

DiffEnc: Variational Diffusion with a Learned Encoder

Beatrix Miranda Ginn Nielsen, Anders Christensen, Andrea Dittadi, Ole Winther
ICLR 2024poster

Diffusion models may be viewed as hierarchical variational autoencoders (VAEs) with two improvements: parameter sharing for the conditionals in the generative process and efficient computation of the loss as independent terms over the hierarchy. We consider two changes to the diffusion model that re…

Cited by 9Source ↗PDF ↗code ↗Scholar ↗
2023

DCI-ES: An Extended Disentanglement Framework with Connections to Identifiability

Cian Eastwood, Andrei Liviu Nicolicioiu, Julius Von Kügelgen, Armin Kekić, Frederik Träuble, Andrea Dittadi +1
ICLR 2023poster

In representation learning, a common approach is to seek representations which disentangle the underlying factors of variation. Eastwood & Williams (2018) proposed three metrics for quantifying the quality of such disentangled representations: disentanglement (D), completeness (C) and informativenes…

Cited by 19Source ↗PDF ↗code ↗Scholar ↗
2022

Assaying Out-Of-Distribution Generalization in Transfer Learning

Florian Wenzel, Andrea Dittadi, Peter Vincent Gehler, Carl-Johann Simon-Gabriel, Max Horn, Dominik Zietlow +6
NeurIPS 2022accept

Since out-of-distribution generalization is a generally ill-posed problem, various proxy targets (e.g., calibration, adversarial robustness, algorithmic corruptions, invariance across shifts) were studied across different research programs resulting in different recommendations. While sharing the sa…

Cited by 80Source ↗PDF ↗code ↗Scholar ↗
2022

Generalization and Robustness Implications in Object-Centric Learning

Andrea Dittadi, Samuele S Papa, Michele De Vita, Bernhard Schölkopf, Ole Winther, Francesco Locatello
ICML 2022spotlight

The idea behind object-centric representation learning is that natural scenes can better be modeled as compositions of objects and their relations as opposed to distributed representations. This inductive bias can be injected into neural networks to potentially improve systematic generalization and…

Cited by 79Source ↗PDF ↗code ↗Scholar ↗
2022

The Role of Pretrained Representations for the OOD Generalization of RL Agents

Frederik Träuble, Andrea Dittadi, Manuel Wuthrich, Felix Widmaier, Peter Vincent Gehler, Ole Winther +4
ICLR 2022poster

Building sample-efficient agents that generalize out-of-distribution (OOD) in real-world settings remains a fundamental unsolved problem on the path towards achieving higher-level cognition. One particularly promising approach is to begin with low-dimensional, pretrained representations of our world…

Cited by 13Source ↗PDF ↗Scholar ↗
2021

Full-Body Motion From a Single Head-Mounted Device: Generating SMPL Poses From Partial Observations

Andrea Dittadi, Sebastian Dziadzio, Darren Cosker, Ben Lundell, Thomas J. Cashman, Jamie Shotton
ICCV 2021poster

The increased availability and maturity of head-mounted and wearable devices opens up opportunities for remote communication and collaboration. However, the signal streams provided by these devices (e.g., head pose, hand pose, and gaze direction) do not represent a whole person. One of the main open…

Cited by 69PDF ↗Scholar ↗
2021

On Disentangled Representations Learned from Correlated Data

Frederik Träuble, Elliot Creager, Niki Kilbertus, Francesco Locatello, Andrea Dittadi, Anirudh Goyal +2
ICML 2021oral

The focus of disentanglement approaches has been on identifying independent factors of variation in data. However, the causal variables underlying real-world observations are often not statistically independent. In this work, we bridge the gap to real-world scenarios by analyzing the behavior of the…

Cited by 142Source ↗PDF ↗code ↗Scholar ↗
2021

On the Transfer of Disentangled Representations in Realistic Settings

Andrea Dittadi, Frederik Träuble, Francesco Locatello, Manuel Wuthrich, Vaibhav Agrawal, Ole Winther +2
ICLR 2021poster

Learning meaningful representations that disentangle the underlying structure of the data generating process is considered to be of key importance in machine learning. While disentangled representations were found to be useful for diverse tasks such as abstract reasoning and fair classification, the…

Cited by 95Source ↗PDF ↗Scholar ↗
2021

Planning from Pixels in Atari with Learned Symbolic Representations

Andrea Dittadi, Frederik K. Drachmann, Thomas Bolander
AAAI 2021technical

Width-based planning methods have been shown to yield state-of-the-art performance in the Atari 2600 domain using pixel input. One successful approach, RolloutIW, represents states with the B-PROST boolean feature set. An augmented version of RolloutIW, pi-IW, shows that learned features can be comp…

Cited by 14Source ↗PDF ↗code ↗Scholar ↗
2020

Optimal Variance Control of the Score-Function Gradient Estimator for Importance-Weighted Bounds

Valentin Liévin, Andrea Dittadi, Anders Christensen, Ole Winther
NeurIPS 2020poster

This paper introduces novel results for the score-function gradient estimator of the importance-weighted variational bound (IWAE). We prove that in the limit of large $K$ (number of importance samples) one can choose the control variate such that the Signal-to-Noise ratio (SNR) of the estimator grow…

Cited by 11Source ↗PDF ↗code ↗Scholar ↗