Paulo Andre Lima de Castro

Gabriel Adriano de Melo, Marcos Ricardo Omena De Albuquerque Maximo, Nei Yoshihiro Soma et al.

The inner alignment problem, which asserts whether an arbitrary artificial intelligence (AI) model satisfices a non-trivial alignment function of its outputs given its inputs, is undecidable. This is rigorously proved by Rice's theorem, which is also equivalent to a reduction to Turing's Halting Problem, whose proof sketch is presented in this work. Nevertheless, there is an enumerable set of provenly aligned AIs that are constructed from a finite set of provenly aligned operations. Therefore, we argue that the alignment should be a guaranteed property from the AI architecture rather than a characteristic imposed post-hoc on an arbitrary AI model. Furthermore, while the outer alignment problem is the definition of a judge function that captures human values and preferences, we propose that such a function must also impose a halting constraint that guarantees that the AI model always reaches a terminal state in finite execution steps. Our work presents examples and models that illustrate this constraint and the intricate challenges involved, advancing a compelling case for adopting an intrinsically hard-aligned approach to AI systems architectures that ensures halting.

Murilo Sibrao Bernardini, Paulo Andre Lima de Castro

In this paper, we propose a method for evaluating autonomous trading strategies that provides realistic expectations, regarding the strategy's long-term performance. This method addresses This method addresses many pitfalls that currently fool even experienced software developers and researchers, not to mention the customers that purchase these products. We present the results of applying our method to several famous autonomous trading strategies, which are used to manage a diverse selection of financial assets. The results show that many of these published strategies are far from being reliable vehicles for financial investment. Our method exposes the difficulties involved in building a reliable, long-term strategy and provides a means to compare potential strategies and select the most promising one by establishing minimal periods and requirements for the test executions. There are many developers that create software to buy and sell financial assets autonomously and some of them present great performance when simulating with historical price series (commonly called backtests). Nevertheless, when these strategies are used in real markets (or data not used in their training or evaluation), quite often they perform very poorly. The proposed method can be used to evaluate potential strategies. In this way, the method helps to tell if you really have a great trading strategy or you are just fooling yourself.