No abstract

Grasping is a complex process involving knowledge of the object, the surroundings, and of oneself. While humans are able to integrate and process all of the sensory information required for performing this task, equipping machines with this capability is an extremely challenging endeavor. In this paper, we investigate how deep learning techniques can allow us to translate high-level concepts such as motor imagery to the problem of robotic grasp synthesis. We explore a paradigm based on generative models for learning integrated object-action representations, and demonstrate its capacity for capturing and generating multimodal, multi-finger grasp configurations on a simulated grasping dataset.

Abstract-Advances in diffuse reflectance infra-red spectroscopy measurements have made it possible to estimate a number of functional properties of soil inexpensively and accurately. Core to such techniques are machine learning methods that can map high-dimensional spectra to real-valued outputs. While previous works have considered predicting each property individually using simple regression methods, the correlation structure present in the output variables prompts us to consider methods that can leverage this structure to make more accurate predictions. In this paper, we leverage advances in deep learning architectures, specifically convolutional neural networks and conditional restricted Boltzmann machines for structured output prediction for soil property prediction. We evaluate our methods on two recent spectral datasets, where output soil properties are shown to have a measurable degree of correlation.

The goal for many PhD students in archaeology is tenure-track employment. Students primarily receive their training by tenure-track or tenured professors, and they are often tacitly expected—or explicitly encouraged—to follow in the footsteps of their advisor. However, the career trajectories that current and recent PhD students follow may hold little resemblance to the ones experienced by their advisors. To understand these different paths and to provide information for current PhD students considering pursuing a career in academia, we surveyed 438 archaeologists holding tenured or tenure-track positions in the United States. The survey, recorded in 2019, posed a variety of questions regarding the personal experiences of individual professors. The results are binned by the decade in which the respondent graduated. Evident patterns are discussed in terms of change over time. The resulting portraits of academic pathways through the past five decades indicate that although broad commonalities exist in the qualifications of early career academics, there is no singular pathway to obtaining tenure-track employment. We highlight the commonalities revealed in our survey to provide a set of general qualifications that might provide a baseline set of skills and experiences for an archaeologist seeking a tenure-track job in the United States.

No abstract