Concept Bottleneck Models (CBMs) are interpretable machine learning models that ground their predictions on human-understandable concepts, allowing for targeted interventions in their decision-making process. However, when intervened on, CBMs assume the availability of humans that can identify the need to intervene and always provide correct interventions. Both assumptions are unrealistic and impractical, considering labor costs and human error-proneness. In contrast, Learning to Defer (L2D) extends supervised learning by allowing machine learning models to identify cases where a human is more likely to be correct than the model, thus leading to deferring systems with improved performance. In this work, we gain inspiration from L2D and propose Deferring CBMs (DCBMs), a novel framework that allows CBMs to learn when an intervention is needed. To this end, we model DCBMs as a composition of deferring systems and derive a consistent L2D loss to train them. Moreover, by relying on a CBM architecture, DCBMs can explain the reasons for deferring on the final task. Our results show that DCBMs can achieve high predictive performance and interpretability by deferring only when needed.

Structure learning is the problem of recovering from data a Directed Acyclic Graph (DAG) of the interactions among variables.By enforcing a differentiable acyclicity constraint on the adjacency matrix of the graph, existing methods solve this problem as an optimization problem and have been recently extended to time-series data.Due to the cubic computational complexity of existing acyclicity constraints, their application is limited to a few variables.In this paper, we introduce svarcosmo, an optimization-based structure learning method for time-series data that builds upon recent developments on unconstrained but provably acyclic models.We empirically show on both simulated and real data that svarcosmo correctly recovers the underlying DAG in significantly less time, enabling optimization-based structure learning on high-dimensional data.

The structure learning problem consists of fitting data generated by a Directed Acyclic Graph (DAG) to correctly reconstruct its arcs. In this context, differentiable approaches constrain or regularize an optimization problem with a continuous relaxation of the acyclicity property. The computational cost of evaluating graph acyclicity is cubic on the number of nodes and significantly affects scalability. In this paper, we introduce COSMO, a constraint-free continuous optimization scheme for acyclic structure learning. At the core of our method lies a novel differentiable approximation of an orientation matrix parameterized by a single priority vector. Differently from previous works, our parameterization fits a smooth orientation matrix and the resulting acyclic adjacency matrix without evaluating acyclicity at any step. Despite this absence, we prove that COSMO always converges to an acyclic solution. In addition to being asymptotically faster, our empirical analysis highlights how COSMO performance on graph reconstruction compares favorably with competing structure learning methods.

The need for modelling causal knowledge at different levels of granularity arises in several settings. Causal Abstraction provides a framework for formalizing this problem by relating two Structural Causal Models at different levels of detail. Despite increasing interest in applying causal abstraction, e.g. in the interpretability of large machine learning models, the graphical and parametrical conditions under which a causal model can abstract another are not known. Furthermore, learning causal abstractions from data is still an open problem. In this work, we tackle both issues for linear causal models with linear abstraction functions. First, we characterize how the low-level coefficients and the abstraction function determine the high-level coefficients and how the high-level model constrains the causal ordering of low-level variables. Then, we apply our theoretical results to learn high-level and low-level causal models and their abstraction function from observational data. In particular, we introduce Abs-LiNGAM, a method that leverages the constraints induced by the learned high-level model and the abstraction function to speedup the recovery of the larger low-level model, under the assumption of non-Gaussian noise terms. In simulated settings, we show the effectiveness of learning causal abstractions from data and the potential of our method in improving scalability of causal discovery.

Highly accurate genotyping is essential for genomic projects aimed at understanding the etiology of diseases as well as for routinary screening of patients. For this reason, genotyping software packages are subject to a strict validation process that requires a large amount of sequencing data endowed with accurate genotype information. In-vitro assessment of genotyping is a long, complex and expensive activity that also depends on the specific variation and locus, and thus it cannot really be used for validation of in-silico genotyping algorithms. In this scenario, sequencing simulation has emerged as a practical alternative. Simulators must be able to keep up with the continuous improvement of different sequencing technologies producing datasets as much indistinguishable from real ones as possible. Moreover, they must be able to mimic as many types of genomic variant as possible. In this paper we describe OmniSim a simulator whose ultimate goal is that of being suitable in all the possible applicative scenarios. In order to fulfill this goal, OmniSim uses an abstract model where variations are read from a .vcf file and mapped into edit operations (insertion, deletion, substitution) on the reference genome. Technological parameters (e.g. error distributions, read length and per-base quality) are learned from real data. As a result of the combination of our abstract model and parameter learning module, OmniSim is able to output data in all aspects similar to that produced in a real sequencing experiment. The source code of OmniSim is freely available at the URL: https://gitlab.com/geraci/omnisim.