A machine learning model was created to predict the electron spectrum generated by a GeV-class laser wakefield accelerator. The model was constructed from variational convolutional neural networks, which mapped the results of secondary laser and plasma diagnostics to the generated electron spectrum. An ensemble of trained networks was used to predict the electron spectrum and to provide an estimation of the uncertainty of that prediction. It is anticipated that this approach will be useful for inferring the electron spectrum prior to undergoing any process that can alter or destroy the beam. In addition, the model provides insight into the scaling of electron beam properties due to stochastic fluctuations in the laser energy and plasma electron density.

Stimulated Raman scattering (SRS) in plasma in a non-eigenmode regime is studied theoretically and numerically. Different from normal SRS with the eigen electrostatic mode excited, the non-eigenmode SRS is developed at plasma density $n_{e}>0.25n_{c}$ when the laser amplitude is larger than a certain threshold. To satisfy the phase-matching conditions of frequency and wavenumber, the excited electrostatic mode has a constant frequency around half of the incident light frequency $\unicode[STIX]{x1D714}_{0}/2$, which is no longer the eigenmode of electron plasma wave $\unicode[STIX]{x1D714}_{pe}$. Both the scattered light and the electrostatic wave are trapped in plasma with their group velocities being zero. Super-hot electrons are produced by the non-eigen electrostatic wave. Our theoretical model is validated by particle-in-cell simulations. The SRS driven in this non-eigenmode regime is an important laser energy loss mechanism in the laser plasma interactions as long as the laser intensity is higher than $10^{15}~\text{W}/\text{cm}^{2}$.

The second and final year of the Erasmus Plus programme ‘Innovative Education and Training in high power laser plasmas’, otherwise known as PowerLaPs, is described. The PowerLaPs programme employs an innovative paradigm in that it is a multi-centre programme, where teaching takes place in five separate institutes with a range of different aims and styles of delivery. The ‘in-class’ time is limited to 4 weeks a year, and the programme spans 2 years. PowerLaPs aims to train students from across Europe in theoretical, applied and laboratory skills relevant to the pursuit of research in laser plasma interaction physics and inertial confinement fusion. Lectures are intermingled with laboratory sessions and continuous assessment activities. The programme, which is led by workers from the Hellenic Mediterranean University and supported by co-workers from the Queen’s University Belfast, the University of Bordeaux, the Czech Technical University in Prague, Ecole Polytechnique, the University of Ioannina, the University of Salamanca and the University of York, has just finished its second and final year. Six Learning Teaching Training activities have been held at the Queen’s University Belfast, the University of Bordeaux, the Czech Technical University, the University of Salamanca and the Institute of Plasma Physics and Lasers of the Hellenic Mediterranean University. The last of these institutes hosted two 2-week-long Intensive Programmes, while the activities at the other four universities were each 5 days in length. In addition, a ‘Multiplier Event’ was held at the University of Ioannina, which will be briefly described. In this second year, the work has concentrated on training in both experimental diagnostics and simulation techniques appropriate to the study of plasma physics, high power laser matter interactions and high energy density physics. The nature of the programme will be described in detail, and some metrics relating to the activities carried out will be presented. In particular, this paper will focus on the overall assessment of the programme.

Absolute instability modes due to secondary scattering of stimulated Raman scattering (SRS) in a large nonuniform plasma are studied theoretically and numerically. The backscattered light of convective SRS can be considered as a pump light with a finite bandwidth. The different frequency components of the backscattered light can be coupled to develop absolute SRS instability near their quarter-critical densities via rescattering process. The absolute SRS mode develops a Langmuir wave with a high phase velocity of about $c/\sqrt{3}$ with $c$ the light speed in vacuum. Given that most electrons are at low velocities in the linear stage, the absolute SRS mode grows with very weak Landau damping. When the interaction evolves into the nonlinear regime, the Langmuir wave can heat abundant electrons up to a few hundred keV via the SRS rescattering. Our theoretical model is validated by particle-in-cell simulations. The absolute instabilities may play a considerable role in the experiments of inertial confinement fusion.