Search results

Improving the science behind the algorithms that run our lives is rapidly becoming one of the most pressing issues of this century. Traditional fixes-laws, regulations and watchdog groups-have proven woefully inadequate. Reporting from the cutting edge of scientific research, The Ethical Algorithm offers a new approach: a set of principled solutions based on the emerging and exciting science of socially aware algorithm design. Michael Kearns and Aaron Roth explain how we can better embed human principles into machine code-without halting the advance of data-driven scientific exploration. The Ethical Algorithm offers a compelling vision for a future, one in which we can better protect humans from the unintended impacts of algorithms while continuing to inspire wondrous advances in technology.

Abstract
Stochastic multiplicative dynamics characterize many complex natural phenomena such as selection and mutation in evolving populations, and the generation and distribution of wealth within social systems. Population heterogeneity in stochastic growth rates has been shown to be the critical driver of wealth inequality over long time scales. However, we still lack a general statistical theory that systematically explains the origins of these heterogeneities resulting from the dynamical adaptation of agents to their environment. In this paper, we derive population growth parameters resulting from the general interaction between agents and their environment, conditional on subjective signals each agent perceives. We show that average wealth-growth rates converge, under specific conditions, to their maximal value as the mutual information between the agent's signal and the environment, and that sequential Bayesian inference is the optimal strategy for reaching this maximum. It follows that when all agents access the same statistical environment, the learning process attenuates growth rate disparities, reducing the long-term effects of heterogeneity on inequality. Our approach shows how the formal properties of information underlie general growth dynamics across social and biological phenomena, including cooperation and the effects of education and learning on life history choices.

We study the design of mechanisms satisfying a novel desideratum: privacy. This requires the mechanism not reveal 'much' about any agent's type to other agents. We propose the notion of joint differential privacy: a variant of differential privacy used in the privacy literature. We show by construction that mechanisms satisfying our desiderata exist when there are a large number of players, and any player's action affects any other's payoff by at most a small amount. Our results imply that in large economies, privacy concerns of agents can be accommodated at no additional 'cost' to standard incentive concerns.

Abstract
Cuneiform is one of the earliest writing systems in recorded human history (ca. 3,400 BCE–75 CE). Hundreds of thousands of such texts were found over the last two centuries, most of which are written in Sumerian and Akkadian. We show the high potential in assisting scholars and interested laypeople alike, by using natural language processing (NLP) methods such as convolutional neural networks (CNN), to automatically translate Akkadian from cuneiform Unicode glyphs directly to English (C2E) and from transliteration to English (T2E). We show that high-quality translations can be obtained when translating directly from cuneiform to English, as we get 36.52 and 37.47 Best Bilingual Evaluation Understudy 4 (BLEU4) scores for C2E and T2E, respectively. For C2E, our model is better than the translation memory baseline in 9.43, and for T2E, the difference is even higher and stands at 13.96. The model achieves best results in short- and medium-length sentences (c. 118 or less characters). As the number of digitized texts grows, the model can be improved by further training as part of a human-in-the-loop system which corrects the results.