A Stability Result for Sparse Convolutions

نویسندگان

  • Philipp Walk
  • Peter Jung
چکیده

We will establish in this note a stability result for sparse convolutions on torsion-free additive (discrete) abelian groups. Sparse convolutions on torsion-free groups are free of cancellations and hence admit stability, i.e. injectivity with a universal lower bound α = α(s, f), only depending on the cardinality s and f of the supports of both input sequences. More precisely, we show that α depends only on s and f and not on the ambient dimension. This statement follows from a reduction argument which involves a compression into a small set preserving the additive structure of the supports.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

On the Stability of Sparse Convolutions

We give a stability result for sparse convolutions on `(G)× `(G) for torsion-free discrete Abelian groups G such as Z. It turns out, that the torsion-free property prevents full cancellation in the convolution of sparse sequences and hence allows to establish stability in each entry, that is, for any fixed entry of the convolution the resulting linear map is injective with an universal lower no...

متن کامل

Polynomials: a new tool for length reduction in binary discrete convolutions

Efficient handling of sparse data is a key challenge in Computer Science. Binary convolutions, such as polynomial multiplication or the Walsh Transform are a useful tool in many applications and are efficiently solved. In the last decade, several problems required efficient solution of sparse binary convolutions. Both randomized and deterministic algorithms were developed for efficiently comput...

متن کامل

On a reverse ℓ2-inequality for sparse circular convolutions

In this paper we show that convolutions of sufficiently sparse signals always admit a non–zero lower bound in energy if oversampling of its Fourier transform is employed. This bound is independent of the signals and the ambient dimension and is determined only be the sparsity of both input signals. This result has several implications for blind system and signal identification and detection, no...

متن کامل

Supplementary Material for Sparsity Invariant CNNs

We find that Sparse Convolutions converge much faster than standard convolutions for most input-output-combinations, especially for those on Synthia with irregularly sparse depth input, as considered in Section 5.1 of the main paper. In Figure 1, we show the mean average error in meters on our validation subset of Synthia over the process of training with identical solver settings (Adam with mo...

متن کامل

Implicit finite difference in time-space domain with the helix transform

Spectral factorization is a method of creating causal filters which have causal inverses. I use spectral factorization of an implicit finite-difference stencil of the two-way wave equation approximation in order to model wave propagation by a sequence of deconvolutions. I deconvolve this filter’s coefficients with the wavefield propagating in a constant velocity medium using the helix approach....

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:
  • CoRR

دوره abs/1312.2222  شماره 

صفحات  -

تاریخ انتشار 2013