192 citations · 334 across the 14 of their papers we have counts for
16 papers
Goal-Conditioned Reinforcement Learning in the Presence of an Adversary
Carlos Purves, Pietro Liò, Cătălina Cangea
Reinforcement learning has seen increasing applications in real-world contexts over the past few years. However, physical environments are often imperfect and policies that perform…
Active Acquisition for Multimodal Temporal Data: A Challenging Decision-Making Task
Jannik Kossen, Cătălina Cangea, Eszter Vértes +5
We introduce a challenging decision-making task that we call active acquisition for multimodal temporal data (A2MT). In many real-world scenarios, input features are not readily av…
General-purpose, long-context autoregressive modeling with Perceiver AR
Curtis Hawthorne, Andrew Jaegle, Cătălina Cangea +12
Real-world data is high-dimensional: a book, image, or musical performance can easily contain hundreds of thousands of elements even after compression. However, the most commonly u…
Structure-aware generation of drug-like molecules
Pavol Drotár, Arian Rokkum Jamasb, Ben Day +2
Structure-based drug design involves finding ligand molecules that exhibit structural and chemical complementarity to protein pockets. Deep generative methods have shown promise in…
Message Passing Neural Processes
Ben Day, Cătălina Cangea, Arian R. Jamasb +1
Neural Processes (NPs) are powerful and flexible models able to incorporate uncertainty when representing stochastic processes, while maintaining a linear time complexity. However,…
Generative Compositional Augmentations for Scene Graph Prediction
Boris Knyazev, Harm de Vries, Cătălina Cangea +3
Inferring objects and their relationships from an image in the form of a scene graph is useful in many applications at the intersection of vision and language. We consider a challe…
Wiki-CS: A Wikipedia-Based Benchmark for Graph Neural Networks
Péter Mernyei, Cătălina Cangea
We present Wiki-CS, a novel dataset derived from Wikipedia for benchmarking Graph Neural Networks. The dataset consists of nodes corresponding to Computer Science articles, with ed…
Sparse Dynamic Distribution Decomposition: Efficient Integration of Trajectory and Snapshot Time Series Data
Jake P. Taylor-King, Cristian Regep, Jyothish Soman +3
Dynamic Distribution Decomposition (DDD) was introduced in Taylor-King et. al. (PLOS Comp Biol, 2020) as a variation on Dynamic Mode Decomposition. In brief, by using basis functio…
Graph Density-Aware Losses for Novel Compositions in Scene Graph Generation
Boris Knyazev, Harm de Vries, Cătălina Cangea +3
Scene graph generation (SGG) aims to predict graph-structured descriptions of input images, in the form of objects and relationships between them. This task is becoming increasingl…
Deep Graph Mapper: Seeing Graphs through the Neural Lens
Cristian Bodnar, Cătălina Cangea, Pietro Liò
Recent advancements in graph representation learning have led to the emergence of condensed encodings that capture the main properties of a graph. However, even though these abstra…
The PlayStation Reinforcement Learning Environment (PSXLE)
Carlos Purves, Cătălina Cangea, Petar Veličković
We propose a new benchmark environment for evaluating Reinforcement Learning (RL) algorithms: the PlayStation Learning Environment (PSXLE), a PlayStation emulator modified to expos…
VideoNavQA: Bridging the Gap between Visual and Embodied Question Answering
Cătălina Cangea, Eugene Belilovsky, Pietro Liò +1
Embodied Question Answering (EQA) is a recently proposed task, where an agent is placed in a rich 3D environment and must act based solely on its egocentric input to answer a given…
Spatio-Temporal Deep Graph Infomax
Felix L. Opolka, Aaron Solomon, Cătălina Cangea +3
Spatio-temporal graphs such as traffic networks or gene regulatory systems present challenges for the existing deep learning methods due to the complexity of structural changes ove…
Structure-Based Networks for Drug Validation
Cătălina Cangea, Arturas Grauslys, Pietro Liò +1
Classifying chemicals according to putative modes of action (MOAs) is of paramount importance in the context of risk assessment. However, current methods are only able to handle a…
Towards Sparse Hierarchical Graph Classifiers
Cătălina Cangea, Petar Veličković, Nikola Jovanović +2
Recent advances in representation learning on graphs, mainly leveraging graph convolutional networks, have brought a substantial improvement on many graph-based benchmark tasks. Wh…
XFlow: Cross-modal Deep Neural Networks for Audiovisual Classification
Cătălina Cangea, Petar Veličković, Pietro Liò
In recent years, there have been numerous developments towards solving multimodal tasks, aiming to learn a stronger representation than through a single modality. Certain aspects o…