BZPEERReading, trust and collaboration for research
Menu

Discover

GraphFollow connections around a paper, topic or saved view.

Workspaces

HomePick up where your research flow left off.InboxHandle requests, replies and notifications from one place.CollectionsCurate reading lists, evidence sets and shareable dossiers.ResearchSee your private provenance graph, trail and imports.CollaborationMove from discovery into focused discussion rooms.PublishDraft, preview and publish full-text research articles.

Network

ExpertsFind specialists worth following or asking for help.CommunitiesJoin research circles organized around shared work.PartnersSee external organizations active around the same topics.

Opportunities

ListingsBrowse roles, calls and collaborations with clearer fit signals.

Account

Log inReturn to your reading and collaboration workspace.Create accountStart saving work, following people and joining teams.
Log inCreate account

Researcher profile

Michael Kopp

Michael Kopp contributes to research discovery and scholarly infrastructure.

ResearcherAffiliation not importedOpen to collaborate
astro-ph.COgr-qchep-phMachine Learningphysics.comp-phastro-ph.IMcond-mat.otherquant-ph

Trust snapshot

Quick read

Trust 21 - EmergingVerification L1Unclaimed author
6works
0followers
8topics
4close collaborators

Actions

Decide how to stay connected

Follow researcher0

Identity and collaboration

How to connect with this researcher

Claiming links this public author record to a researcher profile and unlocks direct collaboration workflows.

Log in to claim

Direct collaboration

Open a focused conversation when the fit is right

Claim this author entity first to unlock direct invitations.

Research graph

See the researcher in context

Open full explorer

Inspect adjacent work, topics, institutions and collaborators without jumping out to a separate graph page.

Building this graph slice

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

6 published item(s)

preprint2022arXiv

Nonclassicality of axion-like dark matter through gravitational self-interactions

Axion-like particles (ALPs) are promising dark matter candidates. They are typically described by a classical field, motivated by large phase space occupation numbers. Here we show that such a description is accompanied by a quantum effect: squeezing due to gravitational self-interactions. For a typical QCD axion today, the onset of squeezing is reached on $\mathrm{μs}$-scales and grows over millennia. Thus within the usual models based on the classical Schrödinger-Poisson equation, a type of Gross-Pitaevskii equation, any viable ALP is nonclassical. We also show that squeezing may be relevant on the scales of other self-gravitating systems such as galactic haloes, or solitonic cores. Conversely, our results highlight the incompleteness and limitations of the classical single field description of ALPs.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Traffic4cast at NeurIPS 2021 -- Temporal and Spatial Few-Shot Transfer Learning in Gridded Geo-Spatial Processes

The IARAI Traffic4cast competitions at NeurIPS 2019 and 2020 showed that neural networks can successfully predict future traffic conditions 1 hour into the future on simply aggregated GPS probe data in time and space bins. We thus reinterpreted the challenge of forecasting traffic conditions as a movie completion task. U-Nets proved to be the winning architecture, demonstrating an ability to extract relevant features in this complex real-world geo-spatial process. Building on the previous competitions, Traffic4cast 2021 now focuses on the question of model robustness and generalizability across time and space. Moving from one city to an entirely different city, or moving from pre-COVID times to times after COVID hit the world thus introduces a clear domain shift. We thus, for the first time, release data featuring such domain shifts. The competition now covers ten cities over 2 years, providing data compiled from over 10^12 GPS probe data. Winning solutions captured traffic dynamics sufficiently well to even cope with these complex domain shifts. Surprisingly, this seemed to require only the previous 1h traffic dynamic history and static road graph as input.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Cross-Domain Few-Shot Learning by Representation Fusion

In order to quickly adapt to new data, few-shot learning aims at learning from few examples, often by using already acquired knowledge. The new data often differs from the previously seen data due to a domain shift, that is, a change of the input-target distribution. While several methods perform well on small domain shifts like new target classes with similar inputs, larger domain shifts are still challenging. Large domain shifts may result in high-level concepts that are not shared between the original and the new domain, whereas low-level concepts like edges in images might still be shared and useful. For cross-domain few-shot learning, we suggest representation fusion to unify different abstraction levels of a deep neural network into one representation. We propose Cross-domain Hebbian Ensemble Few-shot learning (CHEF), which achieves representation fusion by an ensemble of Hebbian learners acting on different layers of a deep neural network. Ablation studies show that representation fusion is a decisive factor to boost cross-domain few-shot learning. On the few-shot datasets miniImagenet and tieredImagenet with small domain shifts, CHEF is competitive with state-of-the-art methods. On cross-domain few-shot benchmark challenges with larger domain shifts, CHEF establishes novel state-of-the-art results in all categories. We further apply CHEF on a real-world cross-domain application in drug discovery. We consider a domain shift from bioactive molecules to environmental chemicals and drugs with twelve associated toxicity prediction tasks. On these tasks, that are highly relevant for computational drug discovery, CHEF significantly outperforms all its competitors. Github: https://github.com/ml-jku/chef

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

On the single classical field description of interacting scalar fields

We test the degree to which interacting Bosonic systems can be approximated by a classical field as total occupation number is increased. This is done with our publicly available code repository, \href{https://github.com/andillio/QIBS}{QIBS}, a massively parallel solver for these systems. We use a number of toy models well studied in the literature and track when the classical field description admits quantum corrections, called the quantum breaktime. This allows us to test claims in the literature regarding the rate of convergence of these systems to the classical evolution. We test a number of initial conditions, including coherent states, number eigenstates, and field number states. We find that of these initial conditions, only number eigenstates do not converge to the classical evolution as occupation number is increased. We find that systems most similar to scalar field dark matter exhibit a logarithmic enhancement in the quantum breaktime with total occupation number. Systems with contact interactions or with field number state initial conditions, and linear dispersions, exhibit a power law enhancement. Finally, we find that the breaktime scaling depends on both model interactions and initial conditions.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Investigating the use of field solvers for simulating classical systems

We explore the use of field solvers as approximations of classical Vlasov-Poisson systems. This correspondence is investigated in both electrostatic and gravitational contexts. We demonstrate the ability of field solvers to be excellent approximations of problems with cold initial condition into the non linear regime. We also investigate extensions of the Schrödinger-Poisson system that employ multiple stacked cold streams, and the von Neumann-Poisson equation as methods that can successfully reproduce the classical evolution of warm initial conditions. We then discuss how appropriate simulation parameters need to be chosen to avoid interference terms, aliasing, and wave behavior in the field solver solutions. We present a series of criteria clarifying how parameters need to be chosen in order to effectively approximate classical solutions.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Using large scale structure data and a halo model to constrain Generalised Dark Matter

Constraints on the properties of the cosmological dark matter have previously been obtained in a model-independent fashion using the Generalised Dark Matter (GDM) framework. Here we extend that work in several directions: We consider the inclusion of WiggleZ matter power spectrum data, and show that this improves the constraints on the two perturbative GDM parameters, $c^2_s$ and $c^2_\text{vis}$, by a factor of 3, for a conservative choice of wavenumber range. A less conservative choice can yield an improvement of up to an order of magnitude compared to previous constraints. In order to examine the robustness of this result we develop a GDM halo model to explore how non-linear structure formation could proceed in this framework, since currently GDM has only been defined perturbatively and only linear theory has been used when generating constraints. We then examine how the halo model affects the constraints obtained from the matter power spectrum data. The less-conservative wavenumber range shows a significant difference between linear and non-linear modelling, with the latter favouring GDM parameters inconsistent with $Λ$CDM, underlining the importance of careful non-linear modelling when using this data. We also use this halo model to establish the robustness of previously obtained constraints, particularly those that involve weak gravitational lensing of the cosmic microwave background. Additionally, we show how the inclusion of neutrino mass as a free parameter affects previous constraints on the GDM parameters.

0 citations0 reviewsNo code linkedOpen access