Higher Rank Substitutions for Tensor Decompositions: Exploring New Dimensions of Gaussian Elimination

A Novel Approach to Tensor Rank Computation and Its Impact on Waring Ranks and Strassen’s Conjecture

Photo by Roman Mager on Unsplash

In the field of computational mathematics, tensor rank computation has gained significant attention for its versatile applications. To tackle this complex issue, the substitution method serves as a higher-dimensional counterpart of Gaussian elimination, setting the stage for exciting advancements and revelations. This article delves into the research presented by Yaroslav Shitov on higher rank substitutions for tensor decompositions, with a particular focus on the direct sum conjecture for Waring ranks and Strassen’s conjecture.

The essence of the substitution method lies in the fact that removing a rank one slice, referred to as ‘s’, and subsequently adding arbitrary scalar multiples of ‘s’ to all other slices in the same direction decreases the minimum rank by precisely one. However, Shitov’s study presents a game-changing way to implement this method, providing new perspectives and implications for both theoretical and applied mathematics.

Shitov outlines a method to embed an initial tensor ‘T’ into a larger linear space, thereby creating room for more complex and nuanced operations. This embedding enables the replacement of a higher rank slice ‘g’ with a family ‘f’ of rank one slices in the new space. An essential aspect to note is that the substitutions performed with respect to ‘g’ in every direction of ‘T’ mirror the effect on the minimum rank as the corresponding substitutions with respect to ‘f’. This parallelism brings a new level of flexibility and adaptability to the process, fostering a more accurate and efficient approach to tensor rank computation.

But how does this new method find practical applications? The answer lies in the resolution of the direct sum conjecture for Waring ranks. Waring ranks have been the subject of extensive research and study within the mathematical community, with their behaviour and properties crucial to a variety of mathematical domains. The ability to effectively manipulate and control these ranks through tensor decomposition has profound implications for the field, and Shitov’s research represents a significant step towards this goal.

Moreover, Shitov’s research presents a potent counterexample to Strassen’s conjecture, a well-known proposal in the realm of tensor rank computation. The established understanding of Strassen’s conjecture is challenged by this work, opening doors to further exploration and debate within the scientific community.

To summarize, the method put forward by Shitov on higher rank substitutions for tensor decompositions serves as a groundbreaking contribution to computational mathematics. It goes beyond the conventional application of Gaussian elimination, introducing innovative ways to calculate tensor ranks. Its implications for Waring ranks and Strassen’s conjecture not only resolve longstanding mathematical queries but also ignite a renewed curiosity for future explorations in the field.

Read full paper heretensor .