Publications

Learn-To-Design: Reinforcement Learning-Assisted Chemical Process Optimization
Eslam G. Al-Sakkari
Ahmed Ragab
Mohamed Ali
Hanane Dagdougui
Daria C. Boffito
Mouloud Amazouz
This paper proposes an AI-assisted approach aimed at accelerating chemical process design through causal incremental reinforcement learning … (voir plus)(CIRL) where an intelligent agent is interacting iteratively with a process simulation environment (e.g., Aspen HYSYS, DWSIM, etc.). The proposed approach is based on an incremental learnable optimizer capable of guiding multi-objective optimization towards optimal design variable configurations, depending on several factors including the problem complexity, selected RL algorithm and hyperparameters tuning. One advantage of this approach is that the agent-simulator interaction significantly reduces the vast search space of design variables, leading to an accelerated and optimized design process. This is a generic causal approach that enables the exploration of new process configurations and provides actionable insights to designers to improve not only the process design but also the design process across various applications. The approach was validated on industrial processes including an absorption-based carbon capture, considering the economic and technological uncertainties of different capture processes, such as energy price, production cost, and storage capacity. It achieved a cost reduction of up to 5.5% for the designed capture process, after a few iterations, while also providing the designer with actionable insights. From a broader perspective, the proposed approach paves the way for accelerating the adoption of decarbonization technologies (CCUS value chains, clean fuel production, etc.) at a larger scale, thus catalyzing climate change mitigation.
2024-07-08
Systems and Control Transactions (publié)
doi.org
All-in-one simulation-based inference
Manuel Gloeckler
Michael Deistler
Christian Dietrich Weilbach
Frank N. Wood
Jakob H. Macke
2024-07-07
Proceedings of the 41st International Conference on Machine Learning (publié)
doi.org
proceedings.mlr.press
AsmDocGen: Generating Functional Natural Language Descriptions for Assembly Code
Jesia Yuki
Mohammadhossein Amouei
Benjamin C. M. Fung
Philippe Charland
Andrew Walenstein
2024-07-07
Proceedings of the 19th International Conference on Software Technologies (publié)
doi.org
Autoformalizing Euclidean Geometry
Logan Murphy
Kaiyu Yang
Jialiang Sun
Zhaoyu Li
Animashree Anandkumar
Xujie Si
Autoformalization involves automatically translating informal math into formal theorems and proofs that are machine-verifiable. Euclidean ge… (voir plus)ometry provides an interesting and controllable domain for studying autoformalization. In this paper, we introduce a neuro-symbolic framework for autoformalizing Euclidean geometry, which combines domain knowledge, SMT solvers, and large language models (LLMs). One challenge in Euclidean geometry is that informal proofs rely on diagrams, leaving gaps in texts that are hard to formalize. To address this issue, we use theorem provers to fill in such diagrammatic information automatically, so that the LLM only needs to autoformalize the explicit textual steps, making it easier for the model. We also provide automatic semantic evaluation for autoformalized theorem statements. We construct LeanEuclid, an autoformalization benchmark consisting of problems from Euclid’s Elements and the UniGeo dataset formalized in the Lean proof assistant. Experiments with GPT-4 and GPT-4V show the capability and limitations of state-of-the-art LLMs on autoformalizing geometry problems. The data and code are available at https://github.com/loganrjmurphy/LeanEuclid.
2024-07-07
Proceedings of the 41st International Conference on Machine Learning (publié)
doi.org
proceedings.mlr.press
Beyond the Norms: Detecting Prediction Errors in Regression Models
Andres Altieri
Marco Romanelli
Georg Pichler
Florence Alberge
Pablo Piantanida
This paper tackles the challenge of detecting unreliable behavior in regression algorithms, which may arise from intrinsic variability (e.g.… (voir plus), aleatoric uncertainty) or modeling errors (e.g., model uncertainty). First, we formally introduce the notion of unreliability in regression, i.e., when the output of the regressor exceeds a specified discrepancy (or error). Then, using powerful tools for probabilistic modeling, we estimate the discrepancy density, and we measure its statistical diversity using our proposed metric for statistical dissimilarity. In turn, this allows us to derive a data-driven score that expresses the uncertainty of the regression outcome. We show empirical improvements in error detection for multiple regression tasks, consistently outperforming popular baseline approaches, and contributing to the broader field of uncertainty quantification and safe machine learning systems.
2024-07-07
Proceedings of the 41st International Conference on Machine Learning (publié)
doi.org
proceedings.mlr.press
CKGConv: General Graph Convolution with Continuous Kernels
Liheng Ma
Soumyasundar Pal
Yitian Zhang
Jiaming Zhou
Yingxue Zhang
Mark J. Coates
The existing definitions of graph convolution, either from spatial or spectral perspectives, are inflexible and not unified. Defining a gene… (voir plus)ral convolution operator in the graph domain is challenging due to the lack of canonical coordinates, the presence of irregular structures, and the properties of graph symmetries. In this work, we propose a novel and general graph convolution framework by parameterizing the kernels as continuous functions of pseudo-coordinates derived via graph positional encoding. We name this Continuous Kernel Graph Convolution (CKGConv). Theoretically, we demonstrate that CKGConv is flexible and expressive. CKGConv encompasses many existing graph convolutions, and exhibits a stronger expressiveness, as powerful as graph transformers in terms of distinguishing non-isomorphic graphs. Empirically, we show that CKGConv-based Networks outperform existing graph convolutional networks and perform comparably to the best graph transformers across a variety of graph datasets. The code and models are publicly available at https://github.com/networkslab/CKGConv.
2024-07-07
Proceedings of the 41st International Conference on Machine Learning (publié)
doi.org
proceedings.mlr.press
Consistent Adversarially Robust Linear Classification: Non-Parametric Setting
Elvis Dopgima Dohmatob
For binary classification in …
2024-07-07
Proceedings of the 41st International Conference on Machine Learning (publié)
proceedings.mlr.press
proceedings.mlr.press
Discovering Environments with XRM
Mohammad Pezeshki
Diane Bouchacourt
Mark Ibrahim
Nicolas Ballas
P Vincent
David Lopez-Paz
Environment annotations are essential for the success of many out-of-distribution (OOD) generalization methods. Unfortunately, these are cos… (voir plus)tly to obtain and often limited by human annotators’ biases. To achieve robust generalization, it is essential to develop algorithms for automatic environment discovery within datasets. Current proposals, which divide examples based on their training error, suffer from one fundamental problem. These methods introduce hyper-parameters and early-stopping criteria, which require a validation set with human-annotated environments, the very information subject to discovery. In this paper, we propose Cross-Risk Minimization (XRM) to address this issue. XRM trains twin networks, each learning from one random half of the training data, while imitating confident held-out mistakes made by its sibling. XRM provides a recipe for hyper-parameter tuning, does not require early-stopping, and can discover environments for all training and validation data. Algorithms built on top of XRM environments achieve oracle worst-group-accuracy, addressing a long-standing challenge in OOD generalization.
2024-07-07
Proceedings of the 41st International Conference on Machine Learning (publié)
proceedings.mlr.press
proceedings.mlr.press
Don't be so Negative! Score-based Generative Modeling with Oracle-assisted Guidance
Saeid Naderiparizi
Xiaoxuan Liang
Berend Zwartsenberg
Setareh Cohan
Frank N. Wood
2024-07-07
Proceedings of the 41st International Conference on Machine Learning (publié)
doi.org
proceedings.mlr.press
EiG-Search: Generating Edge-Induced Subgraphs for GNN Explanation in Linear Time
Shengyao Lu
Bang Liu
Keith G Mills
Jiao He
Di Niu
2024-07-07
Proceedings of the 41st International Conference on Machine Learning (publié)
proceedings.mlr.press
proceedings.mlr.press
Experts Don't Cheat: Learning What You Don't Know By Predicting Pairs
Daniel D. Johnson
Daniel Tarlow
David Duvenaud
Chris J. Maddison
Identifying how much a model …
2024-07-07
Proceedings of the 41st International Conference on Machine Learning (publié)
doi.org
proceedings.mlr.press
Fine-tuning Reinforcement Learning Models is Secretly a Forgetting Mitigation Problem
Maciej Wolczyk
Bartłomiej Cupiał
Mateusz Ostaszewski
Michal Bortkiewicz
Michał Zając
Razvan Pascanu
Lukasz Kuci'nski
Piotr Milo's
Fine-tuning is a widespread technique that allows practitioners to transfer pre-trained capabilities, as recently showcased by the successfu… (voir plus)l applications of foundation models. However, fine-tuning reinforcement learning (RL) models remains a challenge. This work conceptualizes one specific cause of poor transfer, accentuated in the RL setting by the interplay between actions and observations: forgetting of pre-trained capabilities. Namely, a model deteriorates on the state subspace of the downstream task not visited in the initial phase of fine-tuning, on which the model behaved well due to pre-training. This way, we lose the anticipated transfer benefits. We identify conditions when this problem occurs, showing that it is common and, in many cases, catastrophic. Through a detailed empirical analysis of the challenging NetHack and Montezuma’s Revenge environments, we show that standard knowledge retention techniques mitigate the problem and thus allow us to take full advantage of the pre-trained capabilities. In particular, in NetHack, we achieve a new state-of-the-art for neural models, improving the previous best score from
2024-07-07
Proceedings of the 41st International Conference on Machine Learning (publié)
doi.org
proceedings.mlr.press