Mitchell A. Thornton

Jessie M. Henderson, Elena R. Henderson, Clayton A. Harper et al. · oxford

Physically unclonable functions (PUFs) identify integrated circuits using nonlinearly-related challenge-response pairs (CRPs). Ideally, the relationship between challenges and corresponding responses is unpredictable, even if a subset of CRPs is known. Previous work developed a photonic PUF offering improved security compared to non-optical counterparts. Here, we investigate this PUF's susceptibility to Multiple-Valued-Logic-based machine learning attacks. We find that approximately 1,000 CRPs are necessary to train models that predict response bits better than random chance. Given the significant challenge of acquiring a vast number of CRPs from a photonic PUF, our results demonstrate photonic PUF resilience against such attacks.

Andrew Havard, Theodore Manikas, Eric C. Larson et al.

We propose a method to improve steganography by increasing the resilience of stego-media to discovery through steganalysis. Our approach enhances a class of steganographic approaches through the inclusion of a steganographic assistant convolutional neural network (SA-CNN). Previous research showed success in discovering the presence of hidden information within stego-images using trained neural networks as steganalyzers that are applied to stego-images. Our results show that such steganalyzers are less effective when SA-CNN is employed during the generation of a stego-image. We also explore the advantages and disadvantages of representing all the possible outputs of our SA-CNN within a smaller, discrete space, rather than a continuous space. Our SA-CNN enables certain classes of parametric steganographic algorithms to be customized based on characteristics of the cover media in which information is to be embedded. Thus, SA-CNN is adaptive in the sense that it enables the core steganographic algorithm to be especially configured for each particular instance of cover media. Experimental results are provided that employ a recent steganographic technique, S-UNIWARD, both with and without the use of SA-CNN. We then apply both sets of stego-images, those produced with and without SA-CNN, to an exmaple steganalyzer, Yedroudj-Net, and we compare the results. We believe that this approach for the integration of neural networks with hand-crafted algorithms increases the reliability and adaptability of steganographic algorithms.

Mitchell A. Thornton

We develop a class of data-adaptive shrinkage estimators for high-dimensional covariance estimation in which the shrinkage target is a Reynolds projection of the sample covariance under a finite symmetry group selected from a candidate library by held-out predictive performance. The class generalizes the convex shrinkage estimator of Ledoit and Wolf by replacing the scalar-identity target with a structured target derived from a symmetry group when one is available, and generalizes the group-symmetric maximum-likelihood estimator of Shah and Chandrasekaran by combining structural targeting with adaptive convex shrinkage and by selecting the group from data rather than treating it as prespecified. A two-tier procedure performs the group selection: a universal per-candidate evaluation based on held-out negative log-likelihood, optionally preceded by a domain-specific step that constructs the candidate library from structural priors. We establish a finite-sample regret bound for the held-out calibration of the convex combination weight, an oracle inequality for the data-driven group selection, and a quantitative sufficient-match condition under which the proposed estimator dominates Ledoit-Wolf shrinkage in Frobenius mean-squared error. The procedure is illustrated on six real-data problems spanning finance (S&P~500 daily returns), climate (NOAA OISST sea-surface temperature anomalies), genomics (TCGA-BRCA gene expression), radio signal processing (RadioML 2018.A), astronomical imaging (Galaxy10 DECaLS), and natural image patches (CIFAR-10 with a CIFAR-10.1 distribution-shift companion). An empirical comparison is also made against the Bayesian permutation-symmetry estimator of Chojecki and colleagues. Outside the few-shot regime, where structural priors carry the most information per observation, Ledoit-Wolf shrinkage remains the appropriate baseline.

Mitchell A. Thornton

We unify the discrete Fourier transform (DFT), discrete cosine transform (DCT), Walsh-Hadamard, Haar wavelet, Karhunen-Loève transform, and several others along with their continuous counterparts (Fourier transform, Fourier series, spherical harmonics, fractional Fourier transform) under one representation-theoretic principle: each is the eigenbasis of every covariance invariant under a specific finite or compact group, with columns constructed from the irreducible matrix elements of the group via the Peter-Weyl theorem. The unification rests on the Algebraic Diversity (AD) framework, which identifies the matched group of a covariance as the foundational object of second-order signal processing. The data-dependent KLT emerges as the trivial-matched-group limit; classical transforms emerge as the cyclic, dihedral, elementary abelian, iterated wreath, and hybrid wreath cases. Composition rules cover direct, wreath, and semidirect products. The Reed-Muller and arithmetic transforms appear as related change-of-basis transforms on the matched group of Walsh-Hadamard. A polynomial-time algorithm for matched-group discovery, the DAD-CAD relaxation cast as a generalized eigenvalue problem in double-commutator form, closes the operational loop: the matched group of any empirical covariance is discovered without expert judgment, with noise-aware variants via the commutativity residual $δ$ and algebraic coloring index $α$ for finite-SNR settings. The fractional Fourier transform is treated as the metaplectic $SO(2)$ case with Hermite-Gauss matched basis, and a structural principle relates matched group size inversely to transform resolution. Modern applications (massive-MIMO, graph neural networks, transformer attention, point cloud and 3D vision, brain connectivity, single-cell genomics, quantum informatics) are sketched with their matched groups.

Mitchell A. Thornton

We present principles of algebraic diversity (AD), a group-theoretic approach to signal processing exploiting signal symmetry to extract more information per observation, complementing classical methods that use temporal and spatial diversity. The transformations under which a signal's statistics are invariant form a matched group; this group determines the natural transform for analysis, and averaging an estimator over the group action reduces variance without requiring additional snapshots. The viewpoint is broadened in five directions beyond the single-observation measurement of a companion paper. Rank promotion admits AD on scalar data streams and identifies the law of large numbers as the trivial-group case of a $(G, L)$ continuum combining sample-count with group-orbit averaging. An eigentensor hierarchy handles signals with nested symmetry. A blind group-matching methodology identifies the matched group from data via a polynomial-time generalized eigenvalue problem on the unitary Lie algebra, placing the DFT, DCT, and Karhunen--Loève transforms as distinguished points on a transform manifold. A cost-symmetry matching principle then extends AD from measurement to blind and adaptive signal processing generally; blind equalization is the lead detailed example, with the Constant Modulus Algorithm's residual phase ambiguity predicted analytically and matched within $1.6^\circ$ on 3GPP TDL multipath channels, and other blind problems in signal processing are mapped into the framework. Four theorems formalize a structural capacity $κ$, the Rényi-2 analog of Shannon and von Neumann's Rényi-1 entropies, quantifying how a signal's information is organized rather than how much information it contains. AD complements prior algebraic approaches including invariant estimation, minimax robust estimation, algebraic signal processing, and compressed sensing.

Mitchell A. Thornton

We prove that temporal averaging over multiple observations can be replaced by algebraic group action on a single observation for second-order statistical estimation. A General Replacement Theorem establishes conditions under which a group-averaged estimator from one snapshot achieves equivalent subspace decomposition to multi-snapshot covariance estimation, and an Optimality Theorem proves that the symmetric group is universally optimal (yielding the KL transform). The framework unifies the DFT, DCT, and KLT as special cases of group-matched spectral transforms, with a closed-form double-commutator eigenvalue problem for polynomial-time optimal group selection. Five applications are demonstrated: MUSIC DOA estimation from a single snapshot, massive MIMO channel estimation with 64% throughput gain, single-pulse waveform classification at 90% accuracy, graph signal processing with non-Abelian groups, and a new algebraic analysis of transformer LLMs revealing that RoPE uses the wrong algebraic group for 70-80% of attention heads across five models (22,480 head observations), that the optimal group is content-dependent, and that spectral-concentration-based pruning improves perplexity at the 13B scale. All diagnostics require a single forward pass with no gradients or training.

Mitchell A. Thornton

We provide a computable criterion for selecting among Fourier, wavelet, and time-frequency analysis by extending the algebraic diversity (AD) framework to Lie groups acting on $L^2(\mathbb{R})$. To our knowledge, there is no other criterion that provides this selection capability. The group-averaged estimator generalizes from a finite sum over group elements to an integral with respect to Haar measure. A Continuous Replacement Theorem establishes signal-noise separation under equivariance and ergodicity conditions, with a noise operator $\mathcal{N}_G = C_ρ^{-2}$ determined by the Duflo-Moore operator that explains the frequency-dependent noise floor in wavelet analysis as a consequence of the affine group's non-unimodularity. A Unification Theorem shows that classical spectral analysis corresponds to the translation group, wavelet analysis to the affine group, time-frequency analysis to the Heisenberg-Weyl group, and spherical harmonics to SO(3). The commutativity residual $δ$, extended to Hilbert-Schmidt operator norms, provides a principled selection criterion among these groups. A double-commutator generalized eigenvalue problem solves the blind group matching problem in polynomial time. A Discretization Recovery Theorem establishes that all discrete AD results are sampling approximations to the continuous theory, with $\mathbb{Z}_M \to (\mathbb{R},+)$ as $M \to \infty$.

Mitchell A. Thornton

We extend the algebraic diversity (AD) framework from classical signal processing to quantum measurement theory. The central result -- the Quantum Algebraic Diversity (QAD) Theorem -- establishes that a group-structured positive operator-valued measure (POVM) applied to a single copy of a quantum state produces a group-averaged density matrix estimator that recovers the spectral structure of the true density matrix, analogous to the classical result that a group-averaged outer product recovers covariance eigenstructure from a single observation. We establish a formal Classical-Quantum Duality Map connecting classical covariance estimation to quantum state tomography, and prove an Optimality Inheritance Theorem showing that classical group optimality transfers to quantum settings via the Born map. SIC-POVMs are identified as algebraic diversity with the Heisenberg-Weyl group, and mutually unbiased bases (MUBs) as algebraic diversity with the Clifford group, revealing the hierarchy $\mathrm{HW}(d) \subseteq \mathcal{C}(d) \subseteq S_d$ that mirrors the classical hierarchy $\mathbb{Z}_M \subseteq G_{\min} \subseteq S_M$. The double-commutator eigenvalue theorem provides polynomial-time adaptive POVM selection. A worked qubit example demonstrates that the group-averaged estimator from a single Pauli measurement recovers a full-rank approximation to a mixed qubit state, achieving fidelity 0.91 where standard single-basis tomography produces a rank-1 estimate with fidelity 0.71. Monte Carlo simulations on qudits of dimension $d = 2$ through $d = 13$ (200 random states per dimension) confirm that the Heisenberg-Weyl QAD estimator maintains fidelity above 0.90 across all dimensions from a single measurement outcome, while standard tomography fidelity degrades as $\sim 1/d$, with the improvement ratio scaling linearly with $d$ as predicted by the $O(d)$ copy reduction theorem.

Mitchell A. Thornton

The algebraic diversity framework replaces temporal averaging over multiple observations with algebraic group action on a single observation for second-order statistical estimation. The central open problem in this framework is $\textit{group selection}$: given an $M$-dimensional observation with unknown covariance structure, find the finite group whose spectral decomposition best matches the covariance. Naive enumeration of all subgroups of the symmetric group $S_M$ requires exponential time in $M$. We prove that this combinatorial problem reduces to a generalized eigenvalue problem derived from the double commutator of the covariance matrix, yielding a polynomial-time algorithm with complexity $O(d^2M^2 + d^3)$, where $d$ is the dimension of a generator basis. The minimum eigenvector of the double-commutator matrix directly constructs the optimal group generator in closed form, with no iterative optimization. The reduction is exact: the double-commutator minimum eigenvalue is zero if and only if the optimal generator lies in the span of the basis, and its magnitude provides a certifiable optimality gap when it does not. This problem does not appear in the standard catalogs of computational complexity (Garey and Johnson, 1979) and represents a new class linking group theory, matrix analysis, and statistical estimation. We establish connections to independent component analysis (JADE), structured matrix nearness problems, and simultaneous matrix diagonalization, and we show that the double-commutator formulation is the unique approach that is simultaneously polynomial-time, closed-form, and certifiable.