Accurate photometric redshifts are a lynchpin for many future experiments to pin down the cosmological model and for studies of galaxy evolution. In this study, a novel sparse regression framework for photometric redshift estimation is presented. Synthetic data set simulating the Euclid survey and real data from SDSS DR12 are used to train and test the proposed models. We show that approaches which include careful data preparation and model design offer a significant improvement in comparison with several competing machine learning algorithms. Standard implementations of most regression algorithms use the minimization of the sum of squared errors as the objective function. For redshift inference, this induces a bias in the posterior mean of the output distribution, which can be problematic. In this paper, we directly minimize the target metric Δz = (zs − zp)/(1 + zs) and address the bias problem via a distribution-based weighting scheme, incorporated as part of the optimization objective. The results are compared with other machine learning algorithms in the field such as artificial neural networks (ANN), Gaussian processes (GPs) and sparse GPs. The proposed framework reaches a mean absolute Δz = 0.0026(1 + zs), over the redshift range of 0 ≤ zs ≤ 2 on the simulated data, and Δz = 0.0178(1 + zs) over the entire redshift range on the SDSS DR12 survey, outperforming the standard ANNz used in the literature. We also investigate how the relative size of the training sample affects the photometric redshift accuracy. We find that a training sample of >30 per cent of total sample size, provides little additional constraint on the photometric redshifts, and note that our GP formalism strongly outperforms ANNz in the sparse data regime for the simulated data set.
In recent times, the progress of machine learning has facilitated the development of decision support systems that exhibit predictive accuracy, surpassing human capabilities in certain scenarios. However, this improvement has come at the cost of increased model complexity, rendering them black-box models that obscure their internal logic from users. These black boxes are primarily designed to optimize predictive accuracy, limiting their applicability in critical domains such as medicine, law, and finance, where both accuracy and interpretability are crucial factors for model acceptance. Despite the growing body of research on interpretability, there remains a significant dearth of evaluation methods for the proposed approaches. This survey aims to shed light on various evaluation methods employed in interpreting models. Two primary procedures are prevalent in the literature: qualitative and quantitative evaluations. Qualitative evaluations rely on human assessments, while quantitative evaluations utilize computational metrics. Human evaluation commonly manifests as either researcher intuition or well-designed experiments. However, this approach is susceptible to human biases and fatigue and cannot adequately compare two models. Consequently, there has been a recent decline in the use of human evaluation, with computational metrics gaining prominence as a more rigorous method for comparing and assessing different approaches. These metrics are designed to serve specific goals, such as fidelity, comprehensibility, or stability. The existing metrics often face challenges when scaling or being applied to different types of model outputs and alternative approaches. Another important factor that needs to be addressed is that while evaluating interpretability methods, their results may not always be entirely accurate. For instance, relying on the drop in probability to assess fidelity can be problematic, particularly when facing the challenge of out-of-distribution data. Furthermore, a fundamental challenge in the interpretability domain is the lack of consensus regarding its definition and requirements. This issue is compounded in the evaluation process and becomes particularly apparent when assessing comprehensibility.
Understanding the reasoning behind a predictive model’s decision is an important and longstanding problem driven by ethical and legal considerations. Most recent research has focused on the interpretability of supervised models, whereas unsupervised learning has received less attention. However, the majority of the focus was on interpreting the whole model in a manner that undermined accuracy or model assumptions, while local interpretation received much less attention. Therefore, we propose an intrinsic interpretation for the Gaussian mixture model that provides both global insight and local interpretations. We employed the Bhattacharyya coefficient to measure the overlap and divergence across clusters to provide a global interpretation in terms of the differences and similarities between the clusters. By analyzing the GMM exponent with the Garthwaite–Kock corr-max transformation, the local interpretation is provided in terms of the relative contribution of each feature to the overall distance. Experimental results obtained on three datasets show that the proposed interpretation method outperforms the post hoc model-agnostic LIME in determining the feature contribution to the cluster assignment.
Quantum Key Distribution (QKD) enables the sharing of cryptographic keys secured by quantum mechanics. The BB84 protocol assumes single-photon sources, but practical systems rely on weak coherent pulses vulnerable to Photon-Number-Splitting (PNS) attacks. The Gottesman–Lo–Lütkenhaus–Preskill (GLLP) framework addresses these imperfections, deriving secure key rate bounds under limited PNS scenarios. The decoy-state protocol further improves performance by refining single-photon yield estimates, but still considers multi-photon states as insecure, thereby limiting intensities and constraining key rate and distance. More recently, finite-key security bounds for decoy-state QKD have been extended to address general attacks, ensuring security against adversaries capable of exploiting arbitrary strategies. In this work, we focus on a specific class of attacks, the generalized PNS attack, and demonstrate that higher pulse intensities can be securely used by employing Bayesian inference to estimate key parameters directly from observed data. By raising the pulse intensity to 10 photons, we achieve a 50-fold increase in key rate and a 62.2% increase in operational range (about 200 km) compared to the decoy-state protocol. Furthermore, we accurately model after-pulsing using a Hidden Markov Model (HMM) and reveal inaccuracies in decoy-state calculations that may produce erroneous key-rate estimates. While this methodology does not address all possible attacks, it provides a new approach to security proofs in QKD by shifting from worst-case assumption analysis to observation-dependent inference, advancing the reach and efficiency of discrete-variable QKD protocols.
Predicting human behavior has been the subject of many research areas especially in machine learning. Due to its potential benefits, financially or otherwise, researchers have focused on modeling human behavior from recommending items in an online store to predicting the behavior of an entire ecosystem. In this paper, we make an attempt to predict human preference towards natural speech. The proposed approach makes use of extracted user features from the dataset using Singular Value Decomposition (SVD), features extracted from the wave signal using Mel-frequency cepstral coefficients (MFCC) and Radial Basis Function to map the two feature-spaces. The proposed approach was able to reach a Pearson Correlation Coefficient of 0.92 and a 0.258 MAE when compared to the original average scores. The main contribution of the presented work is the fact that mapping the signal-features (MFCC) into an intermediate feature space (SVD) is far more effective than mapping the signal features directly into the desired output. The proposed algorithm outperformed Support Vector Machines (SVM) in all measures, precisely by 88.14% in terms of correlation and by 48.62% in terms of error.