• HOME
  • NEWS
  • EXPLORE
    • CAREER
      • Companies
      • Jobs
    • EVENTS
    • iGEM
      • News
      • Team
    • PHOTOS
    • VIDEO
    • WIKI
  • BLOG
  • COMMUNITY
    • FACEBOOK
    • INSTAGRAM
    • TWITTER
Wednesday, November 18, 2026
BIOENGINEER.ORG
No Result
View All Result
  • Login
  • HOME
  • NEWS
  • EXPLORE
    • CAREER
      • Companies
      • Jobs
        • Lecturer
        • PhD Studentship
        • Postdoc
        • Research Assistant
    • EVENTS
    • iGEM
      • News
      • Team
    • PHOTOS
    • VIDEO
    • WIKI
  • BLOG
  • COMMUNITY
    • FACEBOOK
    • INSTAGRAM
    • TWITTER
  • HOME
  • NEWS
  • EXPLORE
    • CAREER
      • Companies
      • Jobs
        • Lecturer
        • PhD Studentship
        • Postdoc
        • Research Assistant
    • EVENTS
    • iGEM
      • News
      • Team
    • PHOTOS
    • VIDEO
    • WIKI
  • BLOG
  • COMMUNITY
    • FACEBOOK
    • INSTAGRAM
    • TWITTER
No Result
View All Result
Bioengineer.org
No Result
View All Result
Home NEWS Science News Technology

Modified Halpern Proximal Algorithm Solves Multi-Structured Problems in Curved Hadamard Spaces

by
November 18, 2026
in Technology
Reading Time: 7 mins read
0
Modified Halpern Proximal Algorithm Solves Multi-Structured Problems in Curved Hadamard Spaces

Modified Halpern Proximal Algorithm Solves Multi-Structured Problems in Curved Hadamard Spaces

Share on FacebookShare on TwitterShare on LinkedinShare on RedditShare on Telegram

A research team comprising mathematicians Athar Abbas, Somayya Komal, Muhammad Jabir Khan, Hafiz Fukhar-ud-din, and Muhammad Aqeel Ahmad Khan has introduced a novel modified Halpern-type proximal point algorithm designed to resolve multi-structured mathematical problems in nonlinear Hadamard spaces. Published in Fixed Point Theory and Algorithms for Sciences and Engineering, the study resolves unified computational challenges by proving that the perturbed iterative sequence converges strongly to a simultaneous common solution. This common target point resides at the intersection of the zero sets of a finite family of multivalued monotone operators, the solution sets of a finite family of mixed equilibrium problems, and the fixed-point sets of a finite family of nonlinear k-demimetric mappings. In mathematical analysis and applied optimization, establishing strong convergence without requiring linear or linear-like geometric space properties represents an essential theoretical breakthrough. By moving beyond classical Euclidean or linear Hilbert structures into complete geodesic spaces of non-positive curvature, the authors provide a verified algorithmic framework capable of managing complex, intersecting constraints that appear across modern mathematical sciences.

Optimization frameworks fundamentally depend on equilibrium problems and their various generalized extensions, which unify disparate mathematical models under a cohesive theoretical umbrella. The classical formulation originally introduced by Blum and Oettli seeks an element within a nonempty constraint subset such that a designated bifunction remains non-negative for all feasible comparison vectors. This versatile architecture encapsulates minimization problems, variational inequality problems, fixed-point problems, and monotone inclusion problems, while also providing mathematical foundations for economic models, transportation networks, game-theoretic Nash equilibria, complementarity problems, and image reconstruction systems. Building upon these classical foundations, Ceng and Yao formulated the mixed equilibrium problem in Hilbert spaces by incorporating an additional real-valued convex function alongside the governing bifunction. When the auxiliary convex function vanishes identically, the mixed framework reverts directly to the standard equilibrium problem, and when the bifunction vanishes, it reduces to a classical minimization problem. Extending mixed equilibrium problems to non-positive curvature metric geometries required overcoming substantial structural challenges, particularly ensuring that perturbed bifunctions and their associated resolvent operators remain well-defined, unique, monotone, and firmly non-expansive without relying on linear vector spaces.

Monotone operator theory constitutes another indispensable cornerstone of optimization theory, describing physical and economic systems governed by stable equilibrium laws and evolutionary states. A central objective in this analytical discipline is determining a zero point for a multivalued monotone operator, which corresponds to identifying an element in the operator domain where zero is contained in the mapped image set. The solution set of this zero-point inclusion is known to be both closed and convex, encompassing diverse applications such as finding minimizers for proper, lower semicontinuous convex functions, solving differential inclusions, and modeling evolutionary systems across physics, ecology, and mathematical economics. In Euclidean and Hilbert settings, the classical proximal point algorithm initiated by Martinet and generalized by Rockafellar provided an effective iterative mechanism to approximate these zeros. Subsequent developments by Ranjbar, Khatibzadeh, and other researchers generalized proximal point methods to Hadamard spaces through Mann-type and Halpern-type constructions. However, real-world computational formulations often require optimizing systems where monotone operator inclusions operate concurrently with mixed equilibrium constraints and nonlinear mapping transformations, demanding unified iterative methods capable of solving these multi-structured problems simultaneously.

To formulate algorithms outside linear vector environments, mathematicians rely on the geometric properties of geodesic metric spaces and CAT(0) structures. In a metric space equipped with distance function d, a geodesic path between two points represents an isometric distance-preserving mapping from an interval, with its unique image defining a geodesic segment. A metric space is uniquely geodesic if every pair of points is connected by exactly one geodesic segment, and a subset is convex if the entire segment joining any two points lies wholly within the subset. In this geometric setting, a metric space is designated as a CAT(0) space if every geodesic triangle satisfies the comparison inequality formulated by Aleksandrov, meaning distances between points on triangle perimeters are bounded by corresponding distances on comparison triangles in the Euclidean plane. Complete CAT(0) spaces are formally designated as Hadamard spaces, encompassing classical Euclidean spaces, Hilbert spaces, the Hilbert ball, real metric trees, and simply connected Riemannian manifolds with non-positive sectional curvature. Hadamard spaces guarantee Bruhat and Tits’s characteristic inequality, ensuring that midpoints of segments satisfy quadratic norm-like distance bounds essential for establishing iterative convergence.

Analytical operations in Hadamard spaces are facilitated through quasilinearization, which provides a metric analogue to inner products in linear spaces. Through quasilinearization mappings, an ordered pair of points represents a vector, enabling the definition of generalized inner products defined entirely through metric distance combinations. This structure satisfies symmetry, vector addition properties, and the Cauchy-Schwarz inequality, which characterizes geodesically connected CAT(0) spaces. Kakavandi and Amini utilized quasilinearization to construct the formal dual space of a Hadamard space by defining a pseudometric over real scalars and pairs of points, evaluating the supremum of functional Lipschitz variations across distinct space elements. Equivalence classes within this pseudometric space constitute the dual metric space, allowing multivalued operators to map points from a Hadamard space into subsets of its dual space. A multivalued operator is monotone when the dual inner product between mapped dual elements and connecting displacement vectors remains non-negative for all domain elements. For any positive order parameter, the operator resolvent yields a single-valued, firmly non-expansive mapping whose fixed points coincide precisely with the zero solutions of the original monotone operator.

Alongside monotone operators and mixed equilibrium problems, the research team integrated the theory of nonlinear k-demimetric mappings, first generalized to Hadamard spaces by Aremu and colleagues. For a real parameter strictly less than one, a mapping with a nonempty fixed-point set is defined as k-demimetric if the metric quasilinearization between displacement vectors and fixed-point directions is bounded below by a scaled quadratic distance between the point and its transformed image. This expansive category encompasses k-demicontractive mappings and provides closed, convex fixed-point sets. In Hadamard geometries, an averaged transformation combining identity segments and demimetric transformations generates an associated mapping that is quasi-nonexpansive, Δ-demiclosed, and shares an identical fixed-point set with the parent mapping. Furthermore, resolvent operators associated with mixed equilibrium bifunctions and lower semicontinuous convex functions are single-valued, firmly non-expansive, and satisfy quadratic contraction inequalities. Under specific coercivity and compactness assumptions, the domain of these mixed equilibrium resolvents covers the entire Hadamard space, ensuring that the set of fixed points for the resolvent corresponds exactly to the solution set of the mixed equilibrium problem.

The central methodological contribution of the investigation is the formulation of Algorithm 3.1, a modified Halpern-type proximal point scheme designed to simultaneously handle finite families of all three operator classes. Given a closed and convex subset in a Hadamard space, the algorithm generates an iterative sequence starting from arbitrary initial points. At each iteration step, the current iterate is processed through a sequential composition of resolvent operators corresponding to a finite family of multivalued monotone operators, producing an intermediate vector. This intermediate vector is subsequently fed into a composite chain of resolvent operators associated with the finite family of mixed equilibrium problems, yielding a second intermediate state. A convex geodesic combination then fuses an anchor point, the previous iterate, and the mixed equilibrium output using three non-negative parameter sequences that sum to unity. Finally, the next iterative point is generated by forming a weighted geodesic sum of this fused vector and its images under the averaged k-demimetric mappings, governed by strictly positive sequence weights that sum to one across the index set.

To prove that the generated sequence converges strongly to a unique intersection point, the authors developed a three-stage mathematical proof. In the first step, the team established that the sequence of iterates remains uniformly bounded, leveraging the non-expansivity of the composite monotone resolvents and mixed equilibrium resolvents alongside geodesic distance bounds. This uniform boundedness extends directly to all intermediate auxiliary vectors. In the second step, the analysts derived detailed asymptotic limits for several auxiliary difference sequences. Utilizing specialized sequence lemmas, the proof demonstrates that metric distances between successive iterates, composite resolvent evaluations, and transformed demimetric states vanish as iteration indices tend toward infinity. Specifically, the differences between iterates and their intermediate resolvent projections converge to zero, establishing asymptotic regularity across the full family of operators. This systematic deflation ensures that any cluster point of the process must remain invariant under every individual constituent mapping, preventing asymptotic drift across the curved metric manifold.

In the third step of the analytical demonstration, the authors established strong convergence of the sequence to an element in the common intersection set. Because bounded sequences in Hadamard spaces admit Δ-convergent subsequences whose asymptotic centers consist of single points, the authors extracted a convergent subsequence and applied demiclosedness properties. Consequently, the limit point was proven to belong simultaneously to the fixed-point sets of the demimetric mappings, the zero sets of the monotone operators, and the solution sets of the mixed equilibrium problems. Applying metric projection inequalities and quasilinearization limits, the authors then demonstrated that the distance between iterates and the target point converges strictly to zero. The paper also deduces four significant analytical corollaries from this master theorem: Corollary 3.2 establishes convergence when monotone operators are omitted; Corollary 3.3 specializes the system to single operators; Corollary 3.4 addresses minimization, zero-point, and fixed-point problems by setting equilibrium bifunctions to zero; and Corollary 3.5 resolves zero-point and fixed-point problems when both bifunctions and convex functions vanish.

To validate the practical effectiveness and computational speed of the theoretical scheme, the authors performed numerical experiments within the real metric line, defining a two-dimensional multi-structured problem with explicit quadratic and bilinear bifunctions, convex functions, and subdifferential monotone operators. Setting the demimetric parameter to negative one, the experiment evaluated four distinct initial conditions across different starting points and anchor parameters. The numerical tracking monitored error magnitudes, measured as the absolute distance between consecutive iterates against the number of iterations. Across all four test cases, the proposed modified Halpern-type algorithm converged rapidly, outperforming the benchmark algorithm of Moharami and Eskandani as well as the prior scheme of Aremu and collaborators. The experimental trajectory confirmed the theoretical predictions, showing that the iterative sequence reliably locates the common solution point while exhibiting superior convergence speed and robustness, thereby affirming the potential of proximal-type algorithms for resolving multi-structured optimization challenges in curved spaces.

Subject of Research: Modified proximal-type iterative algorithms for multi-structured optimization problems in Hadamard spaces

Article Title: Modified proximal-type algorithms for solving multi-structured problems in Hadamard spaces

Article References: Abbas, A., Komal, S., Khan, M. J., Fukhar-ud-din, H., & Khan, M. A. A. (2026). Modified proximal-type algorithms for solving multi-structured problems in Hadamard spaces. Fixed Point Theory and Algorithms for Sciences and Engineering, 2026(1), Article 26. https://doi.org/10.1186/s13663-026-00843-7

Image Credits: AI Generated

DOI: 10.1186/s13663-026-00843-7

Keywords: Hadamard spaces, CAT(0) spaces, proximal point algorithm, Halpern-type algorithm, mixed equilibrium problems, monotone operators, k-demimetric mappings, strong convergence, geodesic metric spaces, quasilinearization, resolvent operators, fixed point theory, variational inequalities, nonlinear optimization, asymptotic center, Delta-convergence, convex analysis, non-positive curvature

News Source: Denise Maddox. (November 18, 2026). Modified Halpern Proximal Algorithm Solves Multi-Structured Problems in Curved Hadamard Spaces. Scienmag.

Tags: asymptotic centerCAT(0) spacesconvex analysisDelta-convergencefixed point theorygeodesic metric spacesHadamard spacesHalpern-type algorithmk-demimetric mappingsmixed equilibrium problemsmonotone operatorsnon-positive curvaturenonlinear optimizationproximal point algorithmquasilinearizationresolvent operatorsstrong convergencevariational inequalities
Share12Tweet7Share2ShareShareShare1

Related Posts

Decoupled Multiplex Causal Ranking Improves Microservice Root Cause Localization in Benchmarks

Decoupled Multiplex Causal Ranking Improves Microservice Root Cause Localization in Benchmarks

November 18, 2026
Conceptual Framework Models Halogen Surface Treatments for Lightweight High Entropy Alloys

Conceptual Framework Models Halogen Surface Treatments for Lightweight High Entropy Alloys

November 18, 2026

Information-Aware Reduction Preserves Minority Classes in Construction LiDAR Scans

November 18, 2026

Traffic Cordon Policies Mitigate Localized Pollution in Makkah’s Central District

November 18, 2026

POPULAR NEWS

  • Electroacupuncture Mobilizes Unique CD34-Positive Stem-Like Cells in Healthy Dogs

    Electroacupuncture Mobilizes Unique CD34-Positive Stem-Like Cells in Healthy Dogs

    29 shares
    Share 12 Tweet 7
  • Specialists Outline Cervical Cancer Screening Principles for Resource-Limited Settings

    29 shares
    Share 12 Tweet 7
  • Modified Halpern Proximal Algorithm Solves Multi-Structured Problems in Curved Hadamard Spaces

    29 shares
    Share 12 Tweet 7
  • Machine Learning Outperforms Pharmacokinetic Models Predicting Initial Vancomycin Concentrations in Septic Patients

    29 shares
    Share 12 Tweet 7

About

We bring you the latest biotechnology news from best research centers and universities around the world. Check our website.

Follow us

Recent News

Frontal Brainwave Patterns Offer Delirium Clues Beyond Standard Anesthesia Depth Targets

Decoupled Multiplex Causal Ranking Improves Microservice Root Cause Localization in Benchmarks

Pediatric Cobalamin C Deficiency Can Mimic Common Variable Immunodeficiency Features

Subscribe to Blog via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Join 88 other subscribers
  • Contact Us

Bioengineer.org © Copyright 2023 All Rights Reserved.

Welcome Back!

Login to your account below

Forgotten Password?

Retrieve your password

Please enter your username or email address to reset your password.

Log In
No Result
View All Result
  • Homepages
    • Home Page 1
    • Home Page 2
  • News
  • National
  • Business
  • Health
  • Lifestyle
  • Science

Bioengineer.org © Copyright 2023 All Rights Reserved.