Financial institutions face growing pressure to protect customer data while maintaining the ability to obtain business insights through data analysis. To tackle this challenge, Differential Privacy has risen in recent years as a promising solution to guarantee user data privacy. However, the utility of the analytical models trained employing Differential Privacy remains an open question. In this paper, we consider two main approaches: the private-model workflow, employing Centralized Differential Privacy; and the private-data workflow, implementing Local Differential Privacy. We test these two alternatives on an open bank marketing dataset, training classification models and comparing them against a non-private baseline. Findings show a sharp trade-off between privacy and utility in the private-data workflow, which allow for usable models under moderate privacy guarantees (privacy budget in the range 3-5). On the other hand, in the private-model case, adjusting training configurations can help reduce the impact of privacy constraints. Experiments show the sample size, batch size and learning rate to be the most influential hyperparameters, while the noise multiplier and the gradient clipping norm (related to the privacy mechanism) show limited impact. The results provide practical guidance to apply differential privacy in financial environments, supporting both regulatory compliance and effective data use.

In 2019, Gohr introduced neural-distinguishers as a tool to improve differential cryptanalysis using deep learning. Building on this, we propose an Entropy-based Neural Distinguisher for present that requires significantly fewer input bits and network parameters. Our method identifies relevant bit subsets by analyzing the entropy of output differences between ciphertext pairs. We reduce the distinguisher size by: (i) simplifying the network architecture, and (ii) shrinking the input layer via entropy-based bit selection. Our distinguisher achieves accuracy within 1% (resp. 4%) of the state of the art for 7 (resp. 6) rounds of present , using only 28 out of 64 bits and under 10% of the original parameters. Leveraging this efficient setup, we introduce an iterative key recovery method capable of handling 64-bit round keys—unlike Gohr’s 16-bit target. Our approach recovers full keys (64 bits) with 92.8% success (402 out of 433), with all partial recoveries retrieving at least 56 bits and 83.8% retrieving 60 bits or more.

We define a statistical measure of the typical size of words of low weight in a linear code over a finite field. We prove that the dual toric codes coming from polytopes of degree one are characterized, among all dual toric codes, by being extremal with respect to this measure. We also give a geometric interpretation of the minimum distance of dual toric codes and characterize its extremal values. Finally, we obtain exact formulas for the parameters of both primal and dual toric codes associated to polytopes of degree one.

The Goppa Code Distinguishing (GD) problem consists in distinguishing the matrix of a Goppa code from a random matrix. The hardness of this problem is an assumption to prove the security of code-based cryptographic primitives such as McEliece's cryptosystem. Up to now, it is widely believed that the GD problem is a hard decision problem. We present the first method allowing to distinguish alternant and Goppa codes over any field. Our technique can solve the GD problem in polynomial time provided that the codes have sufficiently large rates. The key ingredient is an algebraic characterization of the key-recovery problem. The idea is to consider the rank of a linear system which is obtained by linearizing a particular polynomial system describing a key-recovery attack. It appears that this dimension depends on the type of code considered. Explicit formulas derived from extensive experimentations for the rank are provided for “generic” random, alternant, and Goppa codes over any field. Finally, we give theoretical explanations of these formulas in the case of random codes, alternant codes over any field of characteristic two and binary Goppa codes.

The Goppa Code Distinguishing (GCD) problem consists in distinguishing the matrix of a Goppa code from a random matrix. Up to now, it is widely believed that the GCD problem is a hard decisional problem. We present the first technique allowing to distinguish alternant and Goppa codes over any field. Our technique can solve the GCD problem in polynomial-time provided that the codes have rates sufficiently large. The key ingredient is an algebraic characterization of the key-recovery problem. The idea is to consider the dimension of the solution space of a linearized system deduced from a particular polynomial system describing a key-recovery. It turns out that experimentally this dimension depends on the type of code. Explicit formulas derived from extensive experimentations for the value of the dimension are provided for “generic” random, alternant, and Goppa code over any alphabet. Finally, we give explanations of these formulas in the case of random codes, alternant codes over any field and binary Goppa codes.