Avoiding subtraction and division of stochastic signals using normalizing flows: NFdeconvolve
This addresses a common issue in scientific measurements, such as fluorescence imaging, by enabling noise-robust statistical recovery without subtraction or division, though it is incremental as it applies an existing ML technique to a specific domain.
The paper tackles the problem of noise amplification when subtracting or dividing stochastic signals by using normalizing flows to approximate the probability distribution of the signal of interest, avoiding these operations altogether, and provides an implementation in the NFdeconvolve software package.
Across the scientific realm, we find ourselves subtracting or dividing stochastic signals. For instance, consider a stochastic realization, $x$, generated from the addition or multiplication of two stochastic signals $a$ and $b$, namely $x=a+b$ or $x = ab$. For the $x=a+b$ example, $a$ can be fluorescence background and $b$ the signal of interest whose statistics are to be learned from the measured $x$. Similarly, when writing $x=ab$, $a$ can be thought of as the illumination intensity and $b$ the density of fluorescent molecules of interest. Yet dividing or subtracting stochastic signals amplifies noise, and we ask instead whether, using the statistics of $a$ and the measurement of $x$ as input, we can recover the statistics of $b$. Here, we show how normalizing flows can generate an approximation of the probability distribution over $b$, thereby avoiding subtraction or division altogether. This method is implemented in our software package, NFdeconvolve, available on GitHub with a tutorial linked in the main text.