ML models are built on the assumption that the data used in production will be similar to the data observed in the past, the one that we trained our models on. While this may be true for some specific use cases, most models work in dynamic data environments where data is constantly changing and where "concept drifts" are likely to happen and adversely impact the models' accuracy and reliability. To deal with this, ML models need to be retrained regularly. Or, as stated in Google's "MLOps: Continuous delivery and automation pipelines in machine learning": "To address these challenges and to maintain your model's accuracy in production, you need to do the following: Actively monitor the quality of your model in production [...] and frequently retrain your production models." This concept is called'Continuous Training' (CT) and is part of the MLOps practice. Continuous training seeks to automatically and continuously retrain the model to adapt to changes that might occur in the data.

An Atari-playing artificial intelligence created by researchers at the University of Freiburg in Germany has discovered a never-before-seen bug in the classic game Qbert. Using an inexplicable and seemingly random series of moves, the algorithm achieved an unprecedented high score in a matter of minutes. The researchers explained how they trained their A.I. to achieve an impossible result rivaling James T. Kirk's defeat of the Kobayashi Maru in a paper posted on the preprint side arXiv on February 24. Rather than employing a standard reinforcement learning approach, they used a lesser-known technique called evolutionary strategy. As the name suggests, the method is loosely based of the Darwinian concept of natural selection.

We consider the following sequential allocation process. A benevolent central authority has to allocate a set of indivisible goods to a set of agents whose preferences it is totally ignorant of. We consider the process of allocating objects one after the other by designating an agent and asking her to pick one of the objects among those that remain. The problem consists in choosing the "best" sequence of agents, according to some optimality criterion. We assume that agents have additive preferences over objects. The choice of an optimality criterion depends on three parameters: how utilities of objects are related to their ranking in an agent's preference relation; how the preferences of different agents are correlated; and how social welfare is defined from the agents' utilities. We address the computation of a sequence maximizing expected social welfare under several assumptions. We also address strategical issues.