5 papers
A Reinforcement Learning Environment for Automatic Code Optimization in the MLIR Compiler
Mohammed Tirichine, Nassim Ameur, Nazim Bendib +4
Code optimization is a crucial task that aims to enhance code performance. However, this process is often tedious and complex, highlighting the necessity for automatic code optimiz…
Dependence-Driven, Scalable Quantum Circuit Mapping with Affine Abstractions
Marouane Benbetka, Merwan Bekkar, Riyadh Baghdadi +1
Qubit Mapping is a critical task in Quantum Compilation, as modern Quantum Processing Units (QPUs) are constrained to nearest-neighbor interactions defined by a qubit coupling grap…
Neural Architecture Search with Mixed Bio-inspired Learning Rules
Imane Hamzaoui, Riyadh Baghdadi
Bio-inspired neural networks are attractive for their adversarial robustness, energy frugality, and closer alignment with cortical physiology, yet they often lag behind back-propag…
Sign-Symmetry Learning Rules are Robust Fine-Tuners
Aymene Berriche, Mehdi Zakaria Adjal, Riyadh Baghdadi
Backpropagation (BP) has long been the predominant method for training neural networks due to its effectiveness. However, numerous alternative approaches, broadly categorized under…
Exploring the Knowledge Mismatch Hypothesis: Hallucination Propensity in Small Models Fine-tuned on Data from Larger Models
Phil Wee, Riyadh Baghdadi
Recently, there has been an explosion of large language models created through fine-tuning with data from larger models. These small models able to produce outputs that appear qual…