Search

Diffusion models have achieved unprecedented performance in image generation, yet they suffer from slow inference due to their iterative sampling process. To address this, early-exiting has recently been proposed, where the depth of the denoising network is made adaptive based on the (estimated) difficulty of each sampling step. Here, we discover an interesting "phase transition" in the sampling process of current adaptive diffusion models: the denoising network consistently exits early during the initial sampling steps, until it suddenly switches to utilizing the full network. Based on this, we propose accelerating generation by employing a shallower denoising network in the initial sampling steps and a deeper network in the later steps. We demonstrate empirically that our dual-backbone approach, DuoDiff, outperforms existing early-exit diffusion methods in both inference speed and generation quality. Importantly, DuoDiff is easy to implement and complementary to existing approaches for accelerating diffusion., Comment: Accepted to NeurIPS, see https://openreview.net/forum?id=G7E4tNmmHD

Scaling machine learning models significantly improves their performance. However, such gains come at the cost of inference being slow and resource-intensive. Early-exit neural networks (EENNs) offer a promising solution: they accelerate inference by allowing intermediate layers to exit and produce a prediction early. Yet a fundamental issue with EENNs is how to determine when to exit without severely degrading performance. In other words, when is it 'safe' for an EENN to go 'fast'? To address this issue, we investigate how to adapt frameworks of risk control to EENNs. Risk control offers a distribution-free, post-hoc solution that tunes the EENN's exiting mechanism so that exits only occur when the output is of sufficient quality. We empirically validate our insights on a range of vision and language tasks, demonstrating that risk control can produce substantial computational savings, all the while preserving user-specified performance goals., Comment: 27 pages, 13 figures, 4 tables (incl. appendix)

Legionella are gram-negative, facultative intracellular, and pathogenic bacteria that pose a risk for human health and cause significant energy losses due to extensive preventive heating-up of water installations. We investigate acoustically-driven motion - acoustophoresis - of several Legionella species, Escherichia coli, Pseudomonas aeruginosa, and Acanthamoeba castellanii, a common Legionella host in water. All the investigated cells can be acoustically manipulated in an ultrasonic standing wave in water, as they possess a non-zero acoustic contrast that is positive for all of the cells, leading to the focusing into pressure nodes of the standing wave. Multi-body simulations indicate that an increase in cell concentration could significantly accelerate the rate of focusing due to hydrodynamic interactions - a phenomenon that we call collective hydrodynamic focusing (CHF). The results form a foundation for acoustic manipulation of bacteria and could pave a path towards acoustically-aided detection of Legionella in water.

Current methods for investigation of receptor - ligand interactions in drug discovery are based on three-dimensional complementarity of receptor and ligand surfaces, and they include pharmacophore modelling, QSAR, molecular docking etc. Those methods only consider short-range molecular interactions (distances <5A), and not include long-range interactions (distances >5A) which are essential for kinetic of biochemical reactions because they influence the number of productive collisions between interacting molecules. Previously was shown that the electron-ion interaction potential (EIIP) represents the physical property which determines the long-range properties of biological molecules. This molecular descriptor served as a base for development of the informational spectrum method (ISM), a virtual spectroscopy method for investigation of protein-protein interactions. In this paper, we proposed a new approach to treat small molecules as linear entities, allowing study of the small molecule - protein interaction by ISM. We analyzed here 21 sets of KEGG drug-protein interactions and showed that this new approach allows an efficient discrimination between biologically active and inactive ligands, and consistence with AA regions of their binding site on the target protein., Comment: Keywords: ISM method, CIS spectra, small molecules, smiles notation, target-ligand recognition, protein target regions

Background: The increasing volume and variety of genotypic and phenotypic data is a major defining characteristic of modern biomedical sciences. At the same time, the limitations in technology for generating data and the inherently stochastic nature of biomolecular events have led to the discrepancy between the volume of data and the amount of knowledge gleaned from it. A major bottleneck in our ability to understand the molecular underpinnings of life is the assignment of function to biological macromolecules, especially proteins. While molecular experiments provide the most reliable annotation of proteins, their relatively low throughput and restricted purview have led to an increasing role for computational function prediction. However, accurately assessing methods for protein function prediction and tracking progress in the field remain challenging. Methodology: We have conducted the second Critical Assessment of Functional Annotation (CAFA), a timed challenge to assess computational methods that automatically assign protein function. One hundred twenty-six methods from 56 research groups were evaluated for their ability to predict biological functions using the Gene Ontology and gene-disease associations using the Human Phenotype Ontology on a set of 3,681 proteins from 18 species. CAFA2 featured significantly expanded analysis compared with CAFA1, with regards to data set size, variety, and assessment metrics. To review progress in the field, the analysis also compared the best methods participating in CAFA1 to those of CAFA2. Conclusions: The top performing methods in CAFA2 outperformed the best methods from CAFA1, demonstrating that computational function prediction is improving. This increased accuracy can be attributed to the combined effect of the growing number of experimental annotations and improved methods for function prediction., Comment: Submitted to Genome Biology