SwePub
Sök i SwePub databas

  Utökad sökning

Träfflista för sökning "WFRF:(Gehler M.) "

Sökning: WFRF:(Gehler M.)

  • Resultat 1-3 av 3
Sortera/gruppera träfflistan
   
NumreringReferensOmslagsbildHitta
1.
  •  
2.
  • Soucek, J., et al. (författare)
  • EMC Aspects Of Turbulence Heating Observer (THOR) Spacecraft
  • 2016
  • Ingår i: Proceedings Of 2016 Esa Workshop On Aerospace Emc (Aerospace Emc). - : Institute of Electrical and Electronics Engineers (IEEE). - 9789292213039
  • Konferensbidrag (refereegranskat)abstract
    • Turbulence Heating ObserveR (THOR) is a spacecraft mission dedicated to the study of plasma turbulence in near-Earth space. The mission is currently under study for implementation as a part of ESA Cosmic Vision program. THOR will involve a single spinning spacecraft equipped with state of the art instruments capable of sensitive measurements of electromagnetic fields and plasma particles. The sensitive electric and magnetic field measurements require that the spacecraft-generated emissions are restricted and strictly controlled; therefore a comprehensive EMC program has been put in place already during the study phase. The THOR study team and a dedicated EMC working group are formulating the mission EMC requirements already in the earliest phase of the project to avoid later delays and cost increases related to EMC. This article introduces the THOR mission and reviews the current state of its EMC requirements.
  •  
3.
  • Träuble, F., et al. (författare)
  • The Role Of Pretrained Representations For The Ood Generalization Of Rl Agents
  • 2022
  • Ingår i: ICLR 2022 - 10th International Conference on Learning Representations. - : International Conference on Learning Representations, ICLR.
  • Konferensbidrag (refereegranskat)abstract
    • Building sample-efficient agents that generalize out-of-distribution (OOD) in real-world settings remains a fundamental unsolved problem on the path towards achieving higher-level cognition. One particularly promising approach is to begin with low-dimensional, pretrained representations of our world, which should facilitate efficient downstream learning and generalization. By training 240 representations and over 10,000 reinforcement learning (RL) policies on a simulated robotic setup, we evaluate to what extent different properties of pretrained VAE-based representations affect the OOD generalization of downstream agents. We observe that many agents are surprisingly robust to realistic distribution shifts, including the challenging sim-to-real case. In addition, we find that the generalization performance of a simple downstream proxy task reliably predicts the generalization performance of our RL agents under a wide range of OOD settings. Such proxy tasks can thus be used to select pretrained representations that will lead to agents that generalize. 
  •  
Skapa referenser, mejla, bekava och länka
  • Resultat 1-3 av 3

Kungliga biblioteket hanterar dina personuppgifter i enlighet med EU:s dataskyddsförordning (2018), GDPR. Läs mer om hur det funkar här.
Så här hanterar KB dina uppgifter vid användning av denna tjänst.

 
pil uppåt Stäng

Kopiera och spara länken för att återkomma till aktuell vy