Measuring the distance between machine-produced and human language is a critical open problem. Inspired by empirical findings from psycholinguistics on the periodicity of entropy in language, we propose FACE, a set of metrics based on Fourier Analysis of the estimated Cross-Entropy of language, for measuring the similarity between model-generated and human-written languages. Based on an open-ended generation task and the experimental data from previous studies, we find that FACE can effectively identify the human-model gap, scales with model size, reflects the outcomes of different sampling methods for decoding, correlates well with other evaluation metrics and with human judgment scores.

We propose a new learning framework that captures the tiered structure of many real-world user-interaction applications, where the users can be divided into two groups based on their different tolerance on exploration risks and should be treated separately. In this setting, we simultaneously maintain two policies $\pi^{\text{O}}$ and $\pi^{\text{E}}$: $\pi^{\text{O}}$ (``O'' for ``online'') interacts with more risk-tolerant users from the first tier and minimizes regret by balancing exploration and exploitation as usual, while $\pi^{\text{E}}$ (``E'' for ``exploit'') exclusively focuses on exploitation for risk-averse users from the second tier utilizing the data collected so far. An important question is whether such a separation yields advantages over the standard online setting (i.e., $\pi^{\text{E}}=\pi^{\text{O}}$) for the risk-averse users. We individually consider the gap-independent vs.~gap-dependent settings. For the former, we prove that the separation is indeed not beneficial from a minimax perspective. For the latter, we show that if choosing Pessimistic Value Iteration as the exploitation algorithm to produce $\pi^{\text{E}}$, we can achieve a constant regret for risk-averse users independent of the number of episodes $K$, which is in sharp contrast to the $\Omega(\log K)$ regret for any online RL algorithms in the same setting, while the regret of $\pi^{\text{O}}$ (almost) maintains its online regret optimality and does not need to compromise for the success of $\pi^{\text{E}}$.

When Face Recognition Doesn't Know Your Face Is a Face An estimated 100 million people live with facial differences. As face recognition tech becomes widespread, some say they're getting blocked from accessing essential systems and services. Autumn Gardiner thought updating her driving license would be straightforward. After getting married last year, she headed to the local Department of Motor Vehicles office in Connecticut to get her name changed on her license. While she was there, Gardiner recalls, officials said she needed to update her photo.

The long-run behavior of multi-agent online learning -- and, in particular, no-regret learning -- is relatively well-understood in potential games, where players have common interests. By contrast, in general harmonic games -- the strategic complement of potential games, where players have competing interests -- very little is known outside the narrow subclass of 2 -player zero-sum games with a fully-mixed equilibrium. As a first result, we show that the continuous-time dynamics of FTRL are Poincaré recurrent, i.e., they return arbitrarily close to their starting point infinitely often, and hence fail to converge. In discrete time, the standard, "vanilla" implementation of FTRL may lead to even worse outcomes, eventually trapping the players in a perpetual cycle of best-responses. However, if FTRL is augmented with a suitable extrapolation step -- which includes as special cases the optimistic and mirror-prox variants of FTRL -- we show that learning converges to a Nash equilibrium from any initial condition, and all players are guaranteed at most \mathcal{O}(1) regret.

Eightfold's experience offers insight into the potential and the challenge of applying A.I. to high-stakes decisions like hiring, promoting and charting career paths for workers. The company is at the forefront of using A.I. and data to assess a person's potential for success in a job. That assessment is based on measuring skills and experience rather than on university degrees or personal connections. The skills-based perspective has been embraced by labor market and policy experts as a vehicle for broadening opportunity in America, especially for the nearly two-thirds of workers who do not have four-year college degrees. Screening by degrees hits minority workers particularly hard, eliminating 72 percent of Black adults and 79 percent of Latino adults, compared with 58 percent of non-Hispanic white adults.

It was 6:28 am when I woke up to a text from a friend in Shanghai, China. "Hey, Amanda--is this you?" he wrote via WeChat. I hadn't even had my morning coffee yet. I pulled my phone closer to get a better look. "Yes, it's me," I typed back.

NASA's Perseverance rover has sent back two selfies of its camera-laden'face' and'head' from the Jezero Crater on the surface of Mars. The two snaps show Perseverance's remote sensing mast, which hosts many of the rover's cameras and scientific instruments. They were taken with the SHERLOC WATSON camera, located on the turret at the end of the rover's robotic arm. Perseverance touched down on the Red Planet on February 18 after a nearly seven-month journey through space. It is tasked with seeking traces of fossilised microbial life from Mars' ancient past and to collect rock specimens for return to Earth through future missions to the Red Planet.

We have developed an autonomous robot system that takes well-composed photographs of people at social events, such as weddings and conference receptions. In this article, we outline the overall architecture of the system and describe how the various components interrelate. We also describe our experiences deploying the robot photographer at a number of real-world events. The system is capable of operating in unaltered environments and has been deployed at a number of real-world events. This article gives an overview of the entire robot photographer system, and provides details of the architecture underlying the implementation.

There is a widespread interest among scientists in understanding a specific and well defined form of intelligence, that is human intelligence. For this reason we propose a stronger version of the original Turing test. In particular, we describe here an open-ended set of Turing questions that we are developing at the Center for Brains, Minds, and Machines at MIT -- that is questions about an image. For the Center for Brains, Minds, and Machines the main research goal is the science of intelligence rather than the engineering of intelligence -- the hardware and software of the brain rather than just absolute performance in face identification. Our Turing questions reflect fully these research priorities.

Does Machine Learning Really Work? Does machine learning really work? Over the past decade, machine learning has evolved from a field of laboratory demonstrations to a field of significant commercial value. Machine-learning algorithms have now learned to detect credit card fraud by mining data on past transactions, learned to steer vehicles driving autonomously on public highways at 70 miles an hour, and learned the reading interests of many individuals to assemble personally customized electronic newspapers. A new computational theory of learning is beginning to shed light on fundamental issues, such as the tradeoff among the number of training examples available, the number of hypotheses considered, and the likely accuracy of the learned hypothesis. Newer research is beginning to explore issues such as long-term learning of new representations, the integration of Bayesian inference and induction, and lifelong cumulative learning. This article, based on the keynote talk presented at ...