Showing 1–10 of 10 results for author: Gessner, A

Active learning for affinity prediction of antibodies

Authors: Alexandra Gessner, Sebastian W. Ober, Owen Vickery, Dino Oglić, Talip Uçar

Abstract: The primary objective of most lead optimization campaigns is to enhance the binding affinity of ligands. For large molecules such as antibodies, identifying mutations that enhance antibody affinity is particularly challenging due to the combinatorial explosion of potential mutations. When the structure of the antibody-antigen complex is available, relative binding free energy (RBFE) methods can of… ▽ More The primary objective of most lead optimization campaigns is to enhance the binding affinity of ligands. For large molecules such as antibodies, identifying mutations that enhance antibody affinity is particularly challenging due to the combinatorial explosion of potential mutations. When the structure of the antibody-antigen complex is available, relative binding free energy (RBFE) methods can offer valuable insights into how different mutations will impact the potency and selectivity of a drug candidate, thereby reducing the reliance on costly and time-consuming wet-lab experiments. However, accurately simulating the physics of large molecules is computationally intensive. We present an active learning framework that iteratively proposes promising sequences for simulators to evaluate, thereby accelerating the search for improved binders. We explore different modeling approaches to identify the most effective surrogate model for this task, and evaluate our framework both using pre-computed pools of data and in a realistic full-loop setting. △ Less

Submitted 11 June, 2024; originally announced June 2024.

Spatiotemporal modeling of European paleoclimate using doubly sparse Gaussian processes

Authors: Seth D. Axen, Alexandra Gessner, Christian Sommer, Nils Weitzel, Álvaro Tejero-Cantero

Abstract: Paleoclimatology -- the study of past climate -- is relevant beyond climate science itself, such as in archaeology and anthropology for understanding past human dispersal. Information about the Earth's paleoclimate comes from simulations of physical and biogeochemical processes and from proxy records found in naturally occurring archives. Climate-field reconstructions (CFRs) combine these data int… ▽ More Paleoclimatology -- the study of past climate -- is relevant beyond climate science itself, such as in archaeology and anthropology for understanding past human dispersal. Information about the Earth's paleoclimate comes from simulations of physical and biogeochemical processes and from proxy records found in naturally occurring archives. Climate-field reconstructions (CFRs) combine these data into a statistical spatial or spatiotemporal model. To date, there exists no consensus spatiotemporal paleoclimate model that is continuous in space and time, produces predictions with uncertainty, and can include data from various sources. A Gaussian process (GP) model would have these desired properties; however, GPs scale unfavorably with data of the magnitude typical for building CFRs. We propose to build on recent advances in sparse spatiotemporal GPs that reduce the computational burden by combining variational methods based on inducing variables with the state-space formulation of GPs. We successfully employ such a doubly sparse GP to construct a probabilistic model of European paleoclimate from the Last Glacial Maximum (LGM) to the mid-Holocene (MH) that synthesizes paleoclimate simulations and fossilized pollen proxy data. △ Less

Submitted 15 November, 2022; originally announced November 2022.

Comments: 8 pages, 4 figures, Accepted at 2022 NeurIPS Workshop on Gaussian Processes, Spatiotemporal Modeling, and Decision-making Systems (GPSMDMS)

ProbNum: Probabilistic Numerics in Python

Authors: Jonathan Wenger, Nicholas Krämer, Marvin Pförtner, Jonathan Schmidt, Nathanael Bosch, Nina Effenberger, Johannes Zenn, Alexandra Gessner, Toni Karvonen, François-Xavier Briol, Maren Mahsereci, Philipp Hennig

Abstract: Probabilistic numerical methods (PNMs) solve numerical problems via probabilistic inference. They have been developed for linear algebra, optimization, integration and differential equation simulation. PNMs naturally incorporate prior information about a problem and quantify uncertainty due to finite computational resources as well as stochastic input. In this paper, we present ProbNum: a Python l… ▽ More Probabilistic numerical methods (PNMs) solve numerical problems via probabilistic inference. They have been developed for linear algebra, optimization, integration and differential equation simulation. PNMs naturally incorporate prior information about a problem and quantify uncertainty due to finite computational resources as well as stochastic input. In this paper, we present ProbNum: a Python library providing state-of-the-art probabilistic numerical solvers. ProbNum enables custom composition of PNMs for specific problem classes via a modular design as well as wrappers for off-the-shelf use. Tutorials, documentation, developer guides and benchmarks are available online at www.probnum.org. △ Less

Submitted 3 December, 2021; originally announced December 2021.

High-Dimensional Gaussian Process Inference with Derivatives

Authors: Filip de Roos, Alexandra Gessner, Philipp Hennig

Abstract: Although it is widely known that Gaussian processes can be conditioned on observations of the gradient, this functionality is of limited use due to the prohibitive computational cost of $\mathcal{O}(N^3 D^3)$ in data points $N$ and dimension $D$. The dilemma of gradient observations is that a single one of them comes at the same cost as $D$ independent function evaluations, so the latter are often… ▽ More Although it is widely known that Gaussian processes can be conditioned on observations of the gradient, this functionality is of limited use due to the prohibitive computational cost of $\mathcal{O}(N^3 D^3)$ in data points $N$ and dimension $D$. The dilemma of gradient observations is that a single one of them comes at the same cost as $D$ independent function evaluations, so the latter are often preferred. Careful scrutiny reveals, however, that derivative observations give rise to highly structured kernel Gram matrices for very general classes of kernels (inter alia, stationary kernels). We show that in the low-data regime $N<D$, the Gram matrix can be decomposed in a manner that reduces the cost of inference to $\mathcal{O}(N^2D + (N^2)^3)$ (i.e., linear in the number of dimensions) and, in special cases, to $\mathcal{O}(N^2D + N^3)$. This reduction in complexity opens up new use-cases for inference with gradients especially in the high-dimensional regime, where the information-to-cost ratio of gradient observations significantly increases. We demonstrate this potential in a variety of tasks relevant for machine learning, such as optimization and Hamiltonian Monte Carlo with predictive gradients. △ Less

Submitted 15 February, 2021; originally announced February 2021.

Bayesian Quadrature on Riemannian Data Manifolds

Authors: Christian Fröhlich, Alexandra Gessner, Philipp Hennig, Bernhard Schölkopf, Georgios Arvanitidis

Abstract: Riemannian manifolds provide a principled way to model nonlinear geometric structure inherent in data. A Riemannian metric on said manifolds determines geometry-aware shortest paths and provides the means to define statistical models accordingly. However, these operations are typically computationally demanding. To ease this computational burden, we advocate probabilistic numerical methods for Rie… ▽ More Riemannian manifolds provide a principled way to model nonlinear geometric structure inherent in data. A Riemannian metric on said manifolds determines geometry-aware shortest paths and provides the means to define statistical models accordingly. However, these operations are typically computationally demanding. To ease this computational burden, we advocate probabilistic numerical methods for Riemannian statistics. In particular, we focus on Bayesian quadrature (BQ) to numerically compute integrals over normal laws on Riemannian manifolds learned from data. In this task, each function evaluation relies on the solution of an expensive initial value problem. We show that by leveraging both prior knowledge and an active exploration scheme, BQ significantly reduces the number of required evaluations and thus outperforms Monte Carlo methods on a wide range of integration problems. As a concrete application, we highlight the merits of adopting Riemannian geometry with our proposed framework on a nonlinear dataset from molecular dynamics. △ Less

Submitted 10 June, 2021; v1 submitted 12 February, 2021; originally announced February 2021.

Three-dimensional Models of Core-collapse Supernovae From Low-mass Progenitors With Implications for Crab

Authors: G. Stockinger, H. -Th. Janka, D. Kresse, T. Melson, T. Ertl, M. Gabler, A. Gessner, A. Wongwathanarat, A. Tolstov, S. -C. Leung, K. Nomoto, A. Heger

Abstract: We present 3D full-sphere supernova simulations of non-rotating low-mass (~9 Msun) progenitors, covering the entire evolution from core collapse through bounce and shock revival, through shock breakout from the stellar surface, until fallback is completed several days later. We obtain low-energy explosions [~(0.5-1.0)x 10^{50} erg] of iron-core progenitors at the low-mass end of the core-collapse… ▽ More We present 3D full-sphere supernova simulations of non-rotating low-mass (~9 Msun) progenitors, covering the entire evolution from core collapse through bounce and shock revival, through shock breakout from the stellar surface, until fallback is completed several days later. We obtain low-energy explosions [~(0.5-1.0)x 10^{50} erg] of iron-core progenitors at the low-mass end of the core-collapse supernova (LMCCSN) domain and compare to a super-AGB (sAGB) progenitor with an oxygen-neon-magnesium core that collapses and explodes as electron-capture supernova (ECSN). The onset of the explosion in the LMCCSN models is modelled self-consistently using the Vertex-Prometheus code, whereas the ECSN explosion is modelled using parametric neutrino transport in the Prometheus-HOTB code, choosing different explosion energies in the range of previous self-consistent models. The sAGB and LMCCSN progenitors that share structural similarities have almost spherical explosions with little metal mixing into the hydrogen envelope. A LMCCSN with less 2nd dredge-up results in a highly asymmetric explosion. It shows efficient mixing and dramatic shock deceleration in the extended hydrogen envelope. Both properties allow fast nickel plumes to catch up with the shock, leading to extreme shock deformation and aspherical shock breakout. Fallback masses of <~5x10^{-3} Msun have no significant effects on the neutron star (NS) masses and kicks. The anisotropic fallback carries considerable angular momentum, however, and determines the spin of the newly-born NS. The LMCCSNe model with less 2nd dredge-up results in a hydrodynamic and neutrino-induced NS kick of >40 km/s and a NS spin period of ~30 ms, both not largely different from those of the Crab pulsar at birth. △ Less

Submitted 10 June, 2020; v1 submitted 5 May, 2020; originally announced May 2020.

Comments: 47 pages, 27 figures, 6 tables; minor revisions, accepted by MNRAS

Integrals over Gaussians under Linear Domain Constraints

Authors: Alexandra Gessner, Oindrila Kanjilal, Philipp Hennig

Abstract: Integrals of linearly constrained multivariate Gaussian densities are a frequent problem in machine learning and statistics, arising in tasks like generalized linear models and Bayesian optimization. Yet they are notoriously hard to compute, and to further complicate matters, the numerical values of such integrals may be very small. We present an efficient black-box algorithm that exploits geometr… ▽ More Integrals of linearly constrained multivariate Gaussian densities are a frequent problem in machine learning and statistics, arising in tasks like generalized linear models and Bayesian optimization. Yet they are notoriously hard to compute, and to further complicate matters, the numerical values of such integrals may be very small. We present an efficient black-box algorithm that exploits geometry for the estimation of integrals over a small, truncated Gaussian volume, and to simulate therefrom. Our algorithm uses the Holmes-Diaconis-Ross (HDR) method combined with an analytic version of elliptical slice sampling (ESS). Adapted to the linear setting, ESS allows for rejection-free sampling, because intersections of ellipses and domain boundaries have closed-form solutions. The key idea of HDR is to decompose the integral into easier-to-compute conditional probabilities by using a sequence of nested domains. Remarkably, it allows for direct computation of the logarithm of the integral value and thus enables the computation of extremely small probability masses. We demonstrate the effectiveness of our tailored combination of HDR and ESS on high-dimensional integrals and on entropy search for Bayesian optimization. △ Less

Submitted 2 March, 2020; v1 submitted 21 October, 2019; originally announced October 2019.

Active Multi-Information Source Bayesian Quadrature

Authors: Alexandra Gessner, Javier Gonzalez, Maren Mahsereci

Abstract: Bayesian quadrature (BQ) is a sample-efficient probabilistic numerical method to solve integrals of expensive-to-evaluate black-box functions, yet so far,active BQ learning schemes focus merely on the integrand itself as information source, and do not allow for information transfer from cheaper, related functions. Here, we set the scene for active learning in BQ when multiple related information s… ▽ More Bayesian quadrature (BQ) is a sample-efficient probabilistic numerical method to solve integrals of expensive-to-evaluate black-box functions, yet so far,active BQ learning schemes focus merely on the integrand itself as information source, and do not allow for information transfer from cheaper, related functions. Here, we set the scene for active learning in BQ when multiple related information sources of variable cost (in input and source) are accessible. This setting arises for example when evaluating the integrand requires a complex simulation to be run that can be approximated by simulating at lower levels of sophistication and at lesser expense. We construct meaningful cost-sensitive multi-source acquisition rates as an extension to common utility functions from vanilla BQ (VBQ),and discuss pitfalls that arise from blindly generalizing. Furthermore, we show that the VBQ acquisition policy is a corner-case of all considered cost-sensitive acquisition schemes, which collapse onto one single de-generate policy in the case of one source and constant cost. In proof-of-concept experiments we scrutinize the behavior of our generalized acquisition functions. On an epidemiological model, we demonstrate that active multi-source BQ (AMS-BQ) allocates budget more efficiently than VBQ for learning the integral to a good accuracy. △ Less

Submitted 12 February, 2021; v1 submitted 27 March, 2019; originally announced March 2019.

Petahertz Spintronics

Authors: Florian Siegrist, Julia A. Gessner, Marcus Ossiander, Christian Denker, Yi-** Chang, Malte C. Schroeder, Alexander Guggenmos, Yang Cui, Jakob Walowski, Ulrike Martens, J. K. Dewhurst, Ulf Kleineberg, Markus Muenzenberg, Sangeeta Sharma, Martin Schultze

Abstract: The enigmatic coupling between electronic and magnetic phenomena was one of the riddles propelling the development of modern electromagnetism. Today, the fully controlled electric field evolution of ultrashort laser pulses permits the direct and ultrafast control of electronic properties of matter and is the cornerstone of light-wave electronics. In sharp contrast, because there is no first order… ▽ More The enigmatic coupling between electronic and magnetic phenomena was one of the riddles propelling the development of modern electromagnetism. Today, the fully controlled electric field evolution of ultrashort laser pulses permits the direct and ultrafast control of electronic properties of matter and is the cornerstone of light-wave electronics. In sharp contrast, because there is no first order interaction between light and spins, the magnetic properties of matter can only be affected indirectly on the much slower tens-of-femtosecond timescale in a sequence of optical excitation followed by the rearrangement of the spin structure. Here we record an orders of magnitude faster magnetic switching with sub-femtosecond response time by initiating optical excitations with near-single-cycle laser pulses in a ferromagnetic layer stack. The unfolding dynamics are tracked in real-time by a novel attosecond time-resolved magnetic circular dichroism (atto-MCD) detection scheme revealing optically induced spin and orbital momentum transfer (OISTR) in synchrony with light field driven charge relocation. In tandem with ab-initio quantum dynamical modelling, we show how this mechanism provides simultaneous control over electronic and magnetic properties that are at the heart of spintronic functionality. This first incarnation of attomagnetism observes light field coherent control of spin-dynamics in the initial non-dissipative temporal regime and paves the way towards coherent spintronic applications with Petahertz clock rates. △ Less

Submitted 18 December, 2018; originally announced December 2018.

Comments: 12 pages, 3+1 figures

Hydrodynamical Neutron-star Kicks in Electron-capture Supernovae and Implications for the CRAB Supernova

Authors: Alexandra Gessner, Hans-Thomas Janka

Abstract: Neutron stars (NSs) obtain kicks of typically several 100 km/s at birth. The gravitational tug-boat mechanism can explain these kicks as consequences of asymmetric mass ejection during the supernova (SN) explosion. Support for this hydrodynamic explanation is provided by observations of SN remnants with associated NSs, which confirm the prediction that the bulk of the explosion ejecta, in particul… ▽ More Neutron stars (NSs) obtain kicks of typically several 100 km/s at birth. The gravitational tug-boat mechanism can explain these kicks as consequences of asymmetric mass ejection during the supernova (SN) explosion. Support for this hydrodynamic explanation is provided by observations of SN remnants with associated NSs, which confirm the prediction that the bulk of the explosion ejecta, in particular chemical elements between silicon and the iron group, are dominantly expelled in the hemisphere opposite to the direction of the NS kick. Here, we present a large set of two- and three-dimensional explosion simulations of electron-capture SNe, considering explosion energies between ~3x10^49 erg and ~1.6x10^50 erg. We find that the fast acceleration of the SN shock in the steep density gradient delimiting the O-Ne-Mg core of the progenitor enables such a rapid expansion of neutrino-heated matter that the growth of neutrino-driven convection freezes out quickly in a high-mode spherical harmonics pattern. Since the corresponding momentum asymmetry of the ejecta is very small and the gravitational acceleration by the fast-expanding ejecta abates rapidly, the NS kick velocities are at most a few km/s. The extremely low core compactness of O-Ne-Mg-core progenitors therefore favors hydrodynamic NS kicks much below the ~160 km/s measured for the Crab pulsar. This suggests either that the Crab Nebula is not the remnant of an electron-capture SN, but of a low-mass iron-core progenitor, or that the Crab pulsar was not accelerated by the gravitational tug-boat mechanism but received its kick by a non-hydrodynamic mechanism such as, e.g., anisotropic neutrino emission. △ Less

Submitted 17 August, 2018; v1 submitted 14 February, 2018; originally announced February 2018.

Comments: 23 pages, 9 figures; revised version accounting for referee and reader comments; accepted by ApJ