Abstract: A machine learning model for reinforcement learning uses parameterized families of Markov decision processes (MDP) with latent variables. The system uses latent variables to improve ability of models to transfer knowledge and generalize to new tasks. Accordingly, trained machine learning based models are able to work in unseen environments or combinations of conditions/factors that the machine learning model was never trained on. For example, robots or self-driving vehicles based on the machine learning based models are robust to changing goals and are able to adapt to novel reward functions or tasks flexibly while being able to transfer knowledge about environments and agents to new tasks.

Abstract: The invention concerns a composition comprising at least a first constituent which is an oligosaccharide of formula (I) containing 3 to 5 oside units and having at least a D-galactose unit bound by an ?[1-6] bond with a sucrose unit. The invention is characterized in that in said sucrose unit R1, R2, R3, R4, independently of one another represent a hydrogen atom, an alkyl group containing 1 to 10 carbon atoms and optionally having at least an unsaturation, a sulphate function or else an ose and, at least a second constituent which is a molecule positively charged with a physiological pH and stimulating pinocytosis or a molecule stimulating membrane penetration.

Abstract: The invention concerns a composition comprising at least a first constituent which is an oligosaccharide of formula (I) containing 3 to 5 oside units and having at least a D-galactose unit bound by an &agr;[1-6] bond with a sucrose unit. The invention is characterized in that in said sucrose unit R1, R2, R3, R4, independently of one another represent a hydrogen atom, an alkyl group containing 1 to 10 carbon atoms and optionally having at least an unsaturation, a sulphate function or else an ose and, at least a second constituent which is a molecule positively charged with a physiological pH and stimulating pinocytosis or a molecule stimulating membrane penetration.