Three-dimensional (3D) deep molecular generative models offer the advantage of goal-directed generation based on 3D-dependent properties, such as binding affinity for structure-based design within binding pockets. Traditional benchmarks created to evaluate SMILES or molecular graphs generators, such as GuacaMol or MOSES, are limited to evaluate 3D generators as they do not assess the quality of the generated molecular conformation. In this work, we hence developed GenBench3D, which implements a new benchmark for models producing molecules within a binding pocket. Our main contribution is the Validity3D metric, evaluating the conformation quality using the likelihood of bond lengths and valence angles based on reference values observed in the Cambridge Structural Database. The LiGAN, 3D-SBDD, Pocket2Mol, TargetDiff, DiffSBDD and ResGen models were benchmarked. We show that only between 0% and 11% of generated molecules have valid conformations. Performing local relaxation of generated molecules in the pocket considerably improved the Validity3D for all models by a minimum increase of 40%. For LiGAN, 3D-SBDD, or TargetDiff, the set of valid relaxed molecules shows on average higher Vina score (i.e. worse) than the set of raw generated molecules, indicating that the binding affinity of raw generated molecules might be overestimated. Using the other scoring functions, that give higher importance to ligand strain, only yield improved scores when using valid relaxed molecules. Using valid relaxed molecules, TargetDiff and Pocket2Mol show better median Vina, Glide and Gold PLP scores than other models. We have publicly released GenBench3D on GitHub for broader use: https://github.com/bbaillif/genbench3d
In this paper, we propose a computationally simple estimator of the asymptotic covariance matrix of the Principal Components (PC) factors valid in the presence of cross-correlated idiosyncratic components. The proposed estimator of the asymptotic Mean Square Error (MSE) of PC factors is based on adaptive thresholding the sample covariances of the id iosyncratic residuals with the threshold based on their individual variances. We compare the nite sample performance of condence regions for the PC factors obtained using the proposed asymptotic MSE with those of available extant asymptotic and bootstrap regions and show that the former beats all alternative procedures for a wide variety of idiosyncratic cross-correlation structures.
There is a tight connection between credit access and voting. We show that uncertainty in access to credit pushes voters toward more conservative candidates in US elections. Using a 1% sample of the US population with valid credit reports, we relate access to credit to voting outcomes in all county-by-congressional districts over the period 2004-2016. Specifically, we construct exogenous measures of uncertainty to credit access, i.e. credit score values around which individual total credit amount jumps the most (e.g. around which uncertainty on access to credit is the highest). We then show that a 10pp increase in the share of marginal voters located just around these thresholds increases republican votes by 2.7pp, and reduces that of democrats by 2.6pp. Furthermore, winning candidates in more uncertain constituencies tend to follow a more conservative rhetoric.
We consider a collaborative iterative algorithm with two agents and a fusion center for estimation of a real valued function (or ``model") on the set of real numbers. While the data collected by the agents is private, in every iteration of the algorithm, the models estimated by the agents are uploaded to the fusion center, fused, and, subsequently downloaded by the agents. We consider the estimation spaces at the agents and the fusion center to be Reproducing Kernel Hilbert Spaces (RKHS). Under suitable assumptions on these spaces, we prove that the algorithm is consistent, i.e., there exists a subsequence of the estimated models which converges to a model in the strong topology. To this end, we define estimation operators for the agents, fusion center, and, for every iteration of the algorithm constructively. We define valid input data sequences, study the asymptotic properties of the norm of the estimation operators, and, find sufficient conditions under which the estimation operator until any iteration is uniformly bounded. Using these results, we prove the existence of an estimation operator for the algorithm which implies the consistency of the considered estimation algorithm.
Consider a parameter of interest, which can be consistently estimated under some conditions. Suppose also that we can at least partly test these conditions with specification tests. We consider the common practice of conducting inference on the parameter of interest conditional on not rejecting these tests. We show that if the tested conditions hold, conditional inference is valid, though possibly conservative. This holds generally, without imposing any assumption on the asymptotic dependence between the estimator of the parameter of interest and the specification test.
Conformal prediction provides machine learning models with prediction sets that offer theoretical guarantees, but the underlying assumption of exchangeability limits its applicability to time series data. Furthermore, existing approaches struggle to handle multi-step ahead prediction tasks, where uncertainty estimates across multiple future time points are crucial. We propose JANET (Joint Adaptive predictioN-region Estimation for Time-series), a novel framework for constructing conformal prediction regions that are valid for both univariate and multivariate time series. JANET generalises the inductive conformal framework and efficiently produces joint prediction regions with controlled K-familywise error rates, enabling flexible adaptation to specific application needs. Our empirical evaluation demonstrates JANET's superior performance in multi-step prediction tasks across diverse time series datasets, highlighting its potential for reliable and interpretable uncertainty quantification in sequential data.
One of the most crucial tasks in the future will be to understand what is going on in neural networks, as they will become even more powerful and widely deployed. This work aims to use TDA methods to analyze neural representations. We develop methods for analyzing representations from different architectures and check how one should use them to obtain valid results. Our findings indicate that removing outliers does not have much impact on the results and that we should compare representations with the same number of elements. We applied these methods for ResNet, VGG19, and ViT architectures and found substantial differences along with some similarities. Additionally, we determined that models with similar architecture tend to have a similar topology of representations and models with a larger number of layers change their topology more smoothly. Furthermore, we found that the topology of pre-trained and finetuned models starts to differ in the middle and final layers while remaining quite similar in the initial layers. These findings demonstrate the efficacy of TDA in the analysis of neural network behavior.
Computer-aided synthesis planning (CASP) algorithms have demonstrated expert-level abilities in planning retrosynthetic routes to molecules of low to moderate complexity. However, current search methods assume the sufficiency of reaching arbitrary building blocks, failing to address the common real-world constraint where using specific molecules is desired. To this end, we present a formulation of synthesis planning with starting material constraints. Under this formulation, we propose Double-Ended Synthesis Planning (DESP), a novel CASP algorithm under a bidirectional graph search scheme that interleaves expansions from the target and from the goal starting materials to ensure constraint satisfiability. The search algorithm is guided by a goal-conditioned cost network learned offline from a partially observed hypergraph of valid chemical reactions. We demonstrate the utility of DESP in improving solve rates and reducing the number of search expansions by biasing synthesis planning towards expert goals on multiple new benchmarks. DESP can make use of existing one-step retrosynthesis models, and we anticipate its performance to scale as these one-step model capabilities improve.
Identifying causal effects is a key problem of interest across many disciplines. The two long-standing approaches to estimate causal effects are observational and experimental (randomized) studies. Observational studies can suffer from unmeasured confounding, which may render the causal effects unidentifiable. On the other hand, direct experiments on the target variable may be too costly or even infeasible to conduct. A middle ground between these two approaches is to estimate the causal effect of interest through proxy experiments, which are conducted on variables with a lower cost to intervene on compared to the main target. Akbari et al. [2022] studied this setting and demonstrated that the problem of designing the optimal (minimum-cost) experiment for causal effect identification is NP-complete and provided a naive algorithm that may require solving exponentially many NP-hard problems as a sub-routine in the worst case. In this work, we provide a few reformulations of the problem that allow for designing significantly more efficient algorithms to solve it as witnessed by our extensive simulations. Additionally, we study the closely-related problem of designing experiments that enable us to identify a given effect through valid adjustments sets.
Bayesian causal discovery offers the power to quantify epistemic uncertainties among a broad range of structurally diverse causal theories potentially explaining the data, represented in forms of directed acyclic graphs (DAGs). However, existing methods struggle with efficient DAG sampling due to the complex acyclicity constraint. In this study, we propose a scalable Bayesian approach to effectively learn the posterior distribution over causal graphs given observational data thanks to the ability to generate DAGs without explicitly enforcing acyclicity. Specifically, we introduce a novel differentiable DAG sampling method that can generate a valid acyclic causal graph by mapping an unconstrained distribution of implicit topological orders to a distribution over DAGs. Given this efficient DAG sampling scheme, we are able to model the posterior distribution over causal graphs using a simple variational distribution over a continuous domain, which can be learned via the variational inference framework. Extensive empirical experiments on both simulated and real datasets demonstrate the superior performance of the proposed model compared to several state-of-the-art baselines.
We describe the findings of the fifth Nuanced Arabic Dialect Identification Shared Task (NADI 2024). NADI's objective is to help advance SoTA Arabic NLP by providing guidance, datasets, modeling opportunities, and standardized evaluation conditions that allow researchers to collaboratively compete on pre-specified tasks. NADI 2024 targeted both dialect identification cast as a multi-label task (Subtask~1), identification of the Arabic level of dialectness (Subtask~2), and dialect-to-MSA machine translation (Subtask~3). A total of 51 unique teams registered for the shared task, of whom 12 teams have participated (with 76 valid submissions during the test phase). Among these, three teams participated in Subtask~1, three in Subtask~2, and eight in Subtask~3. The winning teams achieved 50.57 F\textsubscript{1} on Subtask~1, 0.1403 RMSE for Subtask~2, and 20.44 BLEU in Subtask~3, respectively. Results show that Arabic dialect processing tasks such as dialect identification and machine translation remain challenging. We describe the methods employed by the participating teams and briefly offer an outlook for NADI.
We present and evaluate a method called grammar masking, which is used to guide large language models (LLMs) toward producing syntactically correct models for a given context-free grammar. Prompt engineering methods such as few-shot learning or priming can be used to improve the chances of an LLM producing correct syntax, but the more complex the grammar, the more time-consuming and less promising these methods become. Previous work is focused primarily on the usage of either language model training or prompt engineering. In this work, a method is presented that restricts the output to a given grammar using constrained decoding to ensure the output adheres to a valid syntax. We use several DSLs built with MontiCore and task multiple LLMs to produce models with and without constrained decoding. A corresponding parser is used to confirm the syntactic correctness of each model. We show that grammar masking can dramatically improve the modeling capabilities of several LLMs, reducing the need for well-refined prompting while increasing the chance of producing correct models.
In the field of autonomous driving research, the use of immersive virtual reality (VR) techniques is widespread to enable a variety of studies under safe and controlled conditions. However, this methodology is only valid and consistent if the conduct of participants in the simulated setting mirrors their actions in an actual environment. In this paper, we present a first and innovative approach to evaluating what we term the behavioural gap, a concept that captures the disparity in a participant's conduct when engaging in a VR experiment compared to an equivalent real-world situation. To this end, we developed a digital twin of a pre-existed crosswalk and carried out a field experiment (N=18) to investigate pedestrian-autonomous vehicle interaction in both real and simulated driving conditions. In the experiment, the pedestrian attempts to cross the road in the presence of different driving styles and an external Human-Machine Interface (eHMI). By combining survey-based and behavioural analysis methodologies, we develop a quantitative approach to empirically assess the behavioural gap, as a mechanism to validate data obtained from real subjects interacting in a simulated VR-based environment. Results show that participants are more cautious and curious in VR, affecting their speed and decisions, and that VR interfaces significantly influence their actions.
Background: In the absence of prospective data on diabetic foot ulcers (DFU), cross-sectional associations with causal risk factors (peripheral neuropathy, and peripheral arterial disease (PAD)) could be used to establish the validity of plantar thermography for DFU risk stratification. Methods: First, we investigated the associations between the intrinsic clusters of plantar thermographic images with several DFU risk factors using an unsupervised deep-learning framework. We then studied associations between obtained thermography clusters and DFU risk factors. Second, to identify those associations with predictive power, we used supervised learning to train Convolutional Neural Network (CNN) regression/classification models that predicted the risk factor based on the thermograph (and visual) input. Findings: Our dataset comprised 282 thermographs from type 2 diabetes mellitus patients (aged 56.31 +- 9.18 years, 51.42 % males). On clustering, we found two overlapping clusters (silhouette score = 0.10, indicating weak separation). There was strong evidence for associations between assigned clusters and several factors related to diabetic foot ulceration such as peripheral neuropathy, PAD, number of diabetes complications, and composite DFU risk prediction scores such as Martins-Mendes, PODUS-2020, and SIGN. However, models predicting said risk factors had poor performances. Interpretation: The strong associations between intrinsic thermography clusters and several DFU risk factors support the validity of using thermography for characterising DFU risk. However, obtained associations did not prove to be predictive, likely due to, spectrum bias, or because thermography and classical risk factors characterise incompletely overlapping portions of the DFU risk construct. Our findings highlight the challenges in standardising ground truths when defining novel digital biomarkers.
In this paper I present a practical approach for coupling machine learning (ML) algorithms with knowledge bases (KB) ontology formalism. The lack of availability of prior knowledge in dynamic scenarios is without doubt a major barrier for scalable machine intelligence. My view of the interaction between the two tiers intelligence is based on the idea that when knowledge is not readily available at the knowledge base tier, more knowledge can be extracted from the other tier, which has access to trained models from machine learning algorithms. To analyse this hypothesis, I create two experiments based on different datasets, which are related directly to risk-awareness of autonomous systems, analysed by different machine learning algorithms (namely; multi-layer feedforward backpropagation, Naive Bayes, and J48 decision tree). My analysis shows that the two-tiers intelligence approach for coupling ML and KB is computationally valid and the time complexity of the algorithms during the robot mission is linear with the size of the data and knowledge.
Many recent works have explored using language models for planning problems. One line of research focuses on translating natural language descriptions of planning tasks into structured planning languages, such as the planning domain definition language (PDDL). While this approach is promising, accurately measuring the quality of generated PDDL code continues to pose significant challenges. First, generated PDDL code is typically evaluated using planning validators that check whether the problem can be solved with a planner. This method is insufficient because a language model might generate valid PDDL code that does not align with the natural language description of the task. Second, existing evaluation sets often have natural language descriptions of the planning task that closely resemble the ground truth PDDL, reducing the challenge of the task. To bridge this gap, we introduce \benchmarkName, a benchmark designed to evaluate language models' ability to generate PDDL code from natural language descriptions of planning tasks. We begin by creating a PDDL equivalence algorithm that rigorously evaluates the correctness of PDDL code generated by language models by flexibly comparing it against a ground truth PDDL. Then, we present a dataset of $132,037$ text-to-PDDL pairs across 13 different tasks, with varying levels of difficulty. Finally, we evaluate several API-access and open-weight language models that reveal this task's complexity. For example, $87.6\%$ of the PDDL problem descriptions generated by GPT-4o are syntactically parseable, $82.2\%$ are valid, solve-able problems, but only $35.1\%$ are semantically correct, highlighting the need for a more rigorous benchmark for this problem.
Uncertainty quantification of causal effects is crucial for safety-critical applications such as personalized medicine. A powerful approach for this is conformal prediction, which has several practical benefits due to model-agnostic finite-sample guarantees. Yet, existing methods for conformal prediction of causal effects are limited to binary/discrete treatments and make highly restrictive assumptions such as known propensity scores. In this work, we provide a novel conformal prediction method for potential outcomes of continuous treatments. We account for the additional uncertainty introduced through propensity estimation so that our conformal prediction intervals are valid even if the propensity score is unknown. Our contributions are three-fold: (1) We derive finite-sample prediction intervals for potential outcomes of continuous treatments. (2) We provide an algorithm for calculating the derived intervals. (3) We demonstrate the effectiveness of the conformal prediction intervals in experiments on synthetic and real-world datasets. To the best of our knowledge, we are the first to propose conformal prediction for continuous treatments when the propensity score is unknown and must be estimated from data.
Online black-box optimization (BBO) aims to optimize an objective function by iteratively querying a black-box oracle in a sample-efficient way. While prior studies focus on forward approaches such as Gaussian Processes (GPs) to learn a surrogate model for the unknown objective function, they struggle with steering clear of out-of-distribution and invalid designs in scientific discovery tasks. Recently, inverse modeling approaches that map the objective space to the design space with conditional diffusion models have demonstrated impressive capability in learning the data manifold. However, these approaches proceed in an offline fashion with pre-collected data. How to design inverse approaches for online BBO to actively query new data and improve the sample efficiency remains an open question. In this work, we propose Diffusion-BBO, a sample-efficient online BBO framework leveraging the conditional diffusion model as the inverse surrogate model. Diffusion-BBO employs a novel acquisition function Uncertainty-aware Exploration (UaE) to propose scores in the objective space for conditional sampling. We theoretically prove that Diffusion-BBO with UaE achieves a near-optimal solution for online BBO. We also empirically demonstrate that Diffusion-BBO with UaE outperforms existing online BBO baselines across 6 scientific discovery tasks.
We develop a new method for generating prediction sets that combines the flexibility of conformal methods with an estimate of the conditional distribution $P_{Y \mid X}$. Existing methods, such as conformalized quantile regression and probabilistic conformal prediction, usually provide only a marginal coverage guarantee. In contrast, our approach extends these frameworks to achieve approximately conditional coverage, which is crucial for many practical applications. Our prediction sets adapt to the behavior of the predictive distribution, making them effective even under high heteroscedasticity. While exact conditional guarantees are infeasible without assumptions on the underlying data distribution, we derive non-asymptotic bounds that depend on the total variation distance of the conditional distribution and its estimate. Using extensive simulations, we show that our method consistently outperforms existing approaches in terms of conditional coverage, leading to more reliable statistical inference in a variety of applications.
A/B testers that conduct large-scale tests often prioritize lifts as the main outcome metric and want to be able to control costs resulting from false rejections of the null. This work develops a decision-theoretic framework for maximizing profits subject to false discovery rate (FDR) control. We build an empirical Bayes solution for the problem via a greedy knapsack approach. We derive an oracle rule based on ranking the ratio of expected lifts and the cost of wrong rejections using the local false discovery rate (lfdr) statistic. Our oracle decision rule is valid and optimal for large-scale tests. Further, we establish asymptotic validity for the data-driven procedure and demonstrate finite-sample validity in experimental studies. We also demonstrate the merit of the proposed method over other FDR control methods. Finally, we discuss an application to data collected by experiments on the Optimizely platform.