Computational Approaches to the Singer Transfer: Preimages in the Lambda Algebra and $G_k$-Invariant Theory
Abstract
We present a systematic, algorithmic method to compute the preimage of elements under the Singer algebraic transfer. Using the lambda algebra and the invariant-theoretic formula of P.H. Chon and L.M. Ha [5], we formulate the preimage search as a solvable problem in linear algebra. This framework is applied to study key indecomposable elements in the Adams spectral sequence. As a consequence, we show that the proof of the known result that the indecomposable element lies in the image of the fourth Singer transfer, as given by Nguyen Sum in [17], is false. Furthermore, we provide the explicit description of a preimage for the indecomposable element This preimage had not been explicitly determined in the previous work of N.H.V. Hung and V.T.N. Quynh [8]. Finally, our most significant contribution is the construction of a complete \textsc{SageMath} algorithm that fully automates the computation of both the dimension and an explicit basis for the -invariant space . This tool facilitates the verification of our results [11, 12, 13] that were previously computed manually in connection with Singer's conjecture for rank 4.
Keywords
Cite
@article{arxiv.2507.10108,
title = {Computational Approaches to the Singer Transfer: Preimages in the Lambda Algebra and $G_k$-Invariant Theory},
author = {Dang Vo Phuc},
journal= {arXiv preprint arXiv:2507.10108},
year = {2025}
}
Comments
100 pages. The algorithm is optimized to its best possible form. This work represents our dedicated effort to automate the computation of the dimension and an explicit basis for the invariant subspace, which is a challenging problem in algebraic topology. Constructive and thoughtful comments are therefore always welcome