Contextualisation

GammaLearn

• Fosters innovative methods in AI for CTA
• Evaluate the added value of deep learning

Main results (Published)

• Outperforms Hillas+RF on MC and on real data in controlled environment
• But performances on real data could be improved

→ Domain adaptation

Sensitivity of γ-PhysNet to NSB

Simulations and real data discrepencies

Simulations are approximations of the reality

• Poisson approximation
• Independent pixels

• Some pixels are correlated

Data adaptation (or data tuning)

Modify the simulations to fit the acquisitions

Train	Test
Labelled	Labelled
MC+P(λ) ratio=50%/50%	MC+P(λ) ratio=50%/50%

Tab. Dataset composition

Results with data adaptation on Crab (real data)

γ-PhysNet-CBN (Conditional Batch Norm)

Modify the model to make it robust to noise

Setup

Train	Test
Labelled	Labelled
MC+P(λ~Uniform()) ratio=50%/50%	MC+P(λ) ratio=50%/50%

Tab. Dataset composition

Results with multi-modality on simulations

Results with multi-modality on Crab (real data)

Multi-modality

Domain adaptation

Domain adaptation: Set of algorithms and techniques to reduce domain discrepancies

• Take into account unknown differences between the source (labelled, simulations) and target (unlabelled, real data) domains
• Somehow include unlabelled real data in the training (→ label shift)
• Selection, implementation and validation of DANN (focus of this talk), DeepJDOT, DeepCORAL

Comparative study and validation

• Traditional deep learning dataset (digits)

• Most of the difficulty resides in the optimization of the model (conflicting gradients)

  • Automatic determination of the loss coefficients
  • Gradient weighting strategy

• Best performance using DANN paired with Uncertainy Weighting

→ Paper #1: Method-oriented with results on MC (will be send soon for your review)

γ-PhysNet-DANN (Domain Adversarial Neural Network)

Modify the model to make it domain agnostic

Application of γ-PhysNet-DANN to simulations

Conditional domain adaptation

Setup

Train		Test
Source Labelled	Target Unlabelled	Unlabelled
MC ratio=50%/50%	MC+P(λ) ratio~0.1%/99.9% (Label shift)	MC+P(λ) ratio=50%/50%

Tab. Dataset composition

Results with domain adaptation on simulations

Results with domain adaptation on Crab (real data)

Conclusion & Perspectives

• Comparison of Data adaptation, Multi-modality, Domain adaptation to solve simulations vs real data discreprency
• Tested on simulations
• Tested on real data (Crab), both moonlight and no moonlight conditions
• Standard analysis and γ-PhysNet strongly affected by moonlight
• Data adaptation and multi-modality increase the performance in degraded conditions
• The benefits of domain adaptation are not well established yet
• Tuning accurately the MC to data NSB is still the best performing approach
• Domain adaptation can match the best performance on tuned simulations (but more variability)
• γ-PhysNet-CBN with pedestal image conditioning
• Two articles to be published:
• Paper #1: Method-oriented with results on MC (will be send soon for your review)
• Paper #2: Crab analysis
• My PhD ends September 30th