Publications

Wasserstein Distributionally Robust Optimization Through the Lens of Structural Causal Models and Individual Fairness
Ahmad-reza Ehyaei
Golnoosh Farnadi
Samira Samadi
In recent years, Wasserstein Distributionally Robust Optimization (DRO) has garnered substantial interest for its efficacy in data-driven de… (see more)cision-making under distributional uncertainty. However, limited research has explored the application of DRO to address individual fairness concerns, particularly when considering causal structures and sensitive attributes in learning problems. To address this gap, we first formulate the DRO problem from causality and individual fairness perspectives. We then present the DRO dual formulation as an efficient tool to convert the DRO problem into a more tractable and computationally efficient form. Next, we characterize the closed form of the approximate worst-case loss quantity as a regularizer, eliminating the max-step in the min-max DRO problem. We further estimate the regularizer in more general cases and explore the relationship between DRO and classical robust optimization. Finally, by removing the assumption of a known structural causal model, we provide finite sample error bounds when designing DRO with empirical distributions and estimated causal structures to ensure efficiency and robust learning.
2024-12-10
Conference on Neural Information Processing Systems (Accept (poster))
doi.org
openreview.net
When is an Embedder More Promising than Another?
Maxime Darrin
Philippe Formont
Ismail Ben Ayed
Jackie Chi Kit Cheung
Pablo Piantanida
Embedders play a central role in machine learning, projecting any object into numerical representations that can, in turn, be leveraged to p… (see more)erform various downstream tasks. The evaluation of embedding models typically depends on domain-specific empirical approaches utilizing downstream tasks, primarily because of the lack of a standardized framework for comparison. However, acquiring adequately large and representative datasets for conducting these assessments is not always viable and can prove to be prohibitively expensive and time-consuming. In this paper, we present a unified approach to evaluate embedders. First, we establish theoretical foundations for comparing embedding models, drawing upon the concepts of sufficiency and informativeness. We then leverage these concepts to devise a tractable comparison criterion (information sufficiency), leading to a task-agnostic and self-supervised ranking procedure. We demonstrate experimentally that our approach aligns closely with the capability of embedding models to facilitate various downstream tasks in both natural language processing and molecular biology. This effectively offers practitioners a valuable tool for prioritizing model trials.
2024-12-10
Conference on Neural Information Processing Systems (Accept (poster))
doi.org
openreview.net
WorkArena++: Towards Compositional Planning and Reasoning-based Common Knowledge Work Tasks
Léo Boisvert
Megh Thakkar
Maxime Gasse
Massimo Caccia
Thibault Le Sellier De Chezelles
Quentin Cappart
Nicolas Chapados
Alexandre Lacoste
Alexandre Drouin
The ability of large language models (LLMs) to mimic human-like intelligence has led to a surge in LLM-based autonomous agents. Though recen… (see more)t LLMs seem capable of planning and reasoning given user instructions, their effectiveness in applying these capabilities for autonomous task solving remains underexplored. This is especially true in enterprise settings, where automated agents hold the promise of a high impact. To fill this gap, we propose WorkArena++, a novel benchmark consisting of 682 tasks corresponding to realistic workflows routinely performed by knowledge workers. WorkArena++ is designed to evaluate the planning, problem-solving, logical/arithmetic reasoning, retrieval, and contextual understanding abilities of web agents. Our empirical studies across state-of-the-art LLMs and vision-language models (VLMs), as well as human workers, reveal several challenges for such models to serve as useful assistants in the workplace. In addition to the benchmark, we provide a mechanism to effortlessly generate thousands of ground-truth observation/action traces, which can be used for fine-tuning existing models. Overall, we expect this work to serve as a useful resource to help the community progress toward capable autonomous agents. The benchmark can be found at https://github.com/ServiceNow/WorkArena.
2024-12-10
Advances in Neural Information Processing Systems (Accept (Poster))
doi.org
openreview.net
Building on Efficient Foundations: Effective Training of LLMs with Structured Feedforward Layers.
Xiuying Wei
Skander Moalla
Razvan Pascanu
Caglar Gulçehre
2024-12-09
Advances in Neural Information Processing Systems 37 (published)
doi.org
Detecting Brittle Decisions for Free: Leveraging Margin Consistency in Deep Robust Classifiers
Jonas Ngnawe
Sabyasachi Sahoo
Yann Pequignot
Frédéric Precioso
Christian Gagné
Despite extensive research on adversarial training strategies to improve robustness, the decisions of even the most robust deep learning mod… (see more)els can still be quite sensitive to imperceptible perturbations, creating serious risks when deploying them for high-stakes real-world applications. While detecting such cases may be critical, evaluating a model's vulnerability at a per-instance level using adversarial attacks is computationally too intensive and unsuitable for real-time deployment scenarios. The input space margin is the exact score to detect non-robust samples and is intractable for deep neural networks. This paper introduces the concept of margin consistency -- a property that links the input space margins and the logit margins in robust models -- for efficient detection of vulnerable samples. First, we establish that margin consistency is a necessary and sufficient condition to use a model's logit margin as a score for identifying non-robust samples. Next, through comprehensive empirical analysis of various robustly trained models on CIFAR10 and CIFAR100 datasets, we show that they indicate high margin consistency with a strong correlation between their input space margins and the logit margins. Then, we show that we can effectively and confidently use the logit margin to detect brittle decisions with such models. Finally, we address cases where the model is not sufficiently margin-consistent by learning a pseudo-margin from the feature representation. Our findings highlight the potential of leveraging deep representations to assess adversarial vulnerability in deployment scenarios efficiently.
2024-12-09
Advances in Neural Information Processing Systems (published)
doi.org
openreview.net
Interpreting Learned Feedback Patterns in Large Language Models
Amir Abdullah
Rauno Arike
Fazl Barez
David M. Krueger
Luke Marks
Clement Neo
Philip Torr
2024-12-09
Advances in Neural Information Processing Systems 37 (published)
doi.org
openreview.net
Learning Successor Features the Simple Way
Raymond Chua
Arna Ghosh
Christos Kaplanis
Blake Aaron Richards
Doina Precup
In Deep Reinforcement Learning (RL), it is a challenge to learn representations that do not exhibit catastrophic forgetting or interference … (see more)in non-stationary environments. Successor Features (SFs) offer a potential solution to this challenge. However, canonical techniques for learning SFs from pixel-level observations often lead to representation collapse, wherein representations degenerate and fail to capture meaningful variations in the data. More recent methods for learning SFs can avoid representation collapse, but they often involve complex losses and multiple learning phases, reducing their efficiency. We introduce a novel, simple method for learning SFs directly from pixels. Our approach uses a combination of a Temporal-difference (TD) loss and a reward prediction loss, which together capture the basic mathematical definition of SFs. We show that our approach matches or outperforms existing SF learning techniques in both 2D (Minigrid), 3D (Miniworld) mazes and Mujoco, for both single and continual learning scenarios. As well, our technique is efficient, and can reach higher levels of performance in less time than other approaches. Our work provides a new, streamlined technique for learning SFs directly from pixel observations, with no pretraining required.
2024-12-09
Advances in Neural Information Processing Systems 37 (published)
doi.org
openreview.net
Longitudinal reproducibility of brain and spinal cord quantitative MRI biomarkers
Mathieu Boudreau
Agah Karakuzu
Arnaud Boré
Basile Pinsard
Kiril Zelenkovski
Eva Alonso-Ortiz
Julie Boyle
Lune Bellec
Julien Cohen-Adad
Quantitative MRI (qMRI) promises better specificity, accuracy, repeatability, and reproducibility relative to its clinically-used qualitativ… (see more)e MRI counterpart. Longitudinal reproducibility is particularly important in qMRI. The goal is to reliably quantify tissue properties that may be assessed in longitudinal clinical studies throughout disease progression or during treatment. In this work, we present the initial data release of the quantitative MRI portion of the Courtois project on neural modelling (CNeuroMod), where the brain and cervical spinal cord of six participants were scanned at regular intervals over the course of several years. This first release includes 3 years of data collection and up to 10 sessions per participant using quantitative MRI imaging protocols (T1, magnetization transfer (MTR, MTsat), and diffusion). In the brain, T1MP2RAGE, fractional anisotropy (FA), mean diffusivity (MD), and radial diffusivity (RD) all exhibited high longitudinal reproducibility (intraclass correlation coefficient – ICC ≃ 1 and within-subject coefficient of variations – wCV 1%). The spinal cord cross-sectional area (CSA) computed using T2w images and T1MTsatexhibited the best longitudinal reproducibility (ICC ≃ 1 and 0.7 respectively, and wCV 2.4% and 6.9%). Results from this work show the level of longitudinal reproducibility that can be expected from qMRI protocols in the brain and spinal cord in the absence of hardware and software upgrades, and could help in the design of future longitudinal clinical studies.
2024-12-09
Imaging Neuroscience (published)
doi.org
Normalization and effective learning rates in reinforcement learning
Clare Lyle
Zeyu Zheng
Khimya Khetarpal
James Martens
Hado van Hasselt
Razvan Pascanu
Will Dabney
2024-12-09
Advances in Neural Information Processing Systems 37 (published)
doi.org
openreview.net
Repliqa: A Question-Answering Dataset for Benchmarking LLMs on Unseen Reference Content
Joao Monteiro
Pierre-Andre Noel
Étienne Marcotte
Sai Rajeswar
Valentina Zantedeschi
David Vázquez
Nicolas Chapados
Christopher Pal
Perouz Taslakian
Large Language Models (LLMs) are trained on vast amounts of data, most of which is automatically scraped from the internet. This data includ… (see more)es encyclopedic documents that harbor a vast amount of general knowledge (e.g., Wikipedia) but also potentially overlap with benchmark datasets used for evaluating LLMs. Consequently, evaluating models on test splits that might have leaked into the training set is prone to misleading conclusions. To foster sound evaluation of language models, we introduce a new test dataset named RepLiQA, suited for question-answering and topic retrieval tasks. RepLiQA is a collection of five splits of test sets, four of which have not been released to the internet or exposed to LLM APIs prior to this publication. Each sample in RepLiQA comprises (1) a reference document crafted by a human annotator and depicting an imaginary scenario (e.g., a news article) absent from the internet; (2) a question about the document's topic; (3) a ground-truth answer derived directly from the information in the document; and (4) the paragraph extracted from the reference document containing the answer. As such, accurate answers can only be generated if a model can find relevant content within the provided document. We run a large-scale benchmark comprising several state-of-the-art LLMs to uncover differences in performance across models of various types and sizes in a context-conditional language modeling setting. Released splits of RepLiQA can be found here: https://huggingface.co/datasets/ServiceNow/repliqa.
2024-12-09
Advances in Neural Information Processing Systems 37 (published)
doi.org
openreview.net
The State of Data Curation at NeurIPS: An Assessment of Dataset Development Practices in the Datasets and Benchmarks Track
Eshta Bhardwaj
Harshit Gujral
Siyi Wu
Ciara Zogheib
Tegan Maharaj
Christoph Becker
Data curation is a field with origins in librarianship and archives, whose scholarship and thinking on data issues go back centuries, if not… (see more) millennia. The field of machine learning is increasingly observing the importance of data curation to the advancement of both applications and fundamental understanding of machine learning models - evidenced not least by the creation of the Datasets and Benchmarks track itself. This work provides an analysis of dataset development practices at NeurIPS through the lens of data curation. We present an evaluation framework for dataset documentation, consisting of a rubric and toolkit developed through a literature review of data curation principles. We use the framework to assess the strengths and weaknesses in current dataset development practices of 60 datasets published in the NeurIPS Datasets and Benchmarks track from 2021-2023. We summarize key findings and trends. Results indicate greater need for documentation about environmental footprint, ethical considerations, and data management. We suggest targeted strategies and resources to improve documentation in these areas and provide recommendations for the NeurIPS peer-review process that prioritize rigorous data curation in ML. Finally, we provide results in the format of a dataset that showcases aspects of recommended data curation practices. Our rubric and results are of interest for improving data curation practices broadly in the field of ML as well as to data curation and science and technology studies scholars studying practices in ML. Our aim is to support continued improvement in interdisciplinary research on dataset practices, ultimately improving the reusability and reproducibility of new datasets and benchmarks, enabling standardized and informed human oversight, and strengthening the foundation of rigorous and responsible ML research.
2024-12-09
Advances in Neural Information Processing Systems 37 (published)
doi.org
openreview.net
BootsTAP: Bootstrapped Training for Tracking-Any-Point
Carl Doersch
Yi Yang
Dilara Gokay
Pauline Luc
Skanda Koppula
Ankush Gupta
Joseph Heyward
Ignacio Rocco
Ross Goroshin
João Carreira
Andrew Zisserman
To endow models with greater understanding of physics and motion, it is useful to enable them to perceive how solid surfaces move and deform… (see more) in real scenes. This can be formalized as Tracking-Any-Point (TAP), which requires the algorithm to track any point on solid surfaces in a video, potentially densely in space and time. Large-scale groundtruth training data for TAP is only available in simulation, which currently has a limited variety of objects and motion. In this work, we demonstrate how large-scale, unlabeled, uncurated real-world data can improve a TAP model with minimal architectural changes, using a selfsupervised student-teacher setup. We demonstrate state-of-the-art performance on the TAP-Vid benchmark surpassing previous results by a wide margin: for example, TAP-Vid-DAVIS performance improves from 61.3% to 67.4%, and TAP-Vid-Kinetics from 57.2% to 62.5%. For visualizations, see our project webpage at https://bootstap.github.io/
2024-12-07
Lecture Notes in Computer Science (published)
doi.org
arxiv.org