More Like this

1. Surveillance agriculture and peasant autonomy

   https://doi.org/10.1111/joac.12470  

   Stone, Glenn Davis (July 2022, Journal of Agrarian Change)

   Abstract ‘Big Data’ digital technologies are beginning to make inroads into peasant agriculture in the Global South. Of particular importance is the subset of technologies that appropriate agricultural decision‐making, here theorized as surveillance agriculture. These technological regimes aspire to not only remove decision‐making from the farmer, but eventually to replace the farmer with, for instance, ‘autonomous’ tractors. This paper looks ahead to ask what a technological trajectory that aspires to autonomy for the tractor may portend for autonomy for the peasant farmer. It compares surveillance agriculture to other forms of surveillance capitalism, noting that it shares a will to not only sell products and services but to manipulate behaviour but differs in that the behaviour being manipulated is professional productive behaviour. The paper surveys the vested interests of the entities behind surveillance agriculture and asks how informational relations of production may be changed between farmers and these entities. It then examines the informational relations of production among peasant farmers that may be interdicted by surveillance agriculture, especially the group‐level decision‐making dynamics that make ‘individual autonomy’ a misnomer. But surveillance agriculture is resolutely individualized, which raises concerns for peasant decision‐making autonomy. 

   more » « less
2. Changing notions of teacher autonomy: The intersection of teacher autonomy and instructional improvement in the US

   https://doi.org/10.1177/00345237211055843  

   Lyle, Angela M; Peurach, Donald J (January 2022, Research in Education)

   Historically, teachers had been delegated the primary responsibility for the organization and management of classroom instruction in US public schools. While this delegation afforded teachers professional autonomy in their work, it has also resulted in disparities in students’ educational experiences and outcomes within and between classrooms, schools, and systems. In the effort to improve instruction and reduce disparities for students on a large scale, one reform effort in the US has focused on building instructionally focused education systems (IFESs) where central office and school leaders collaborate with teachers to organize and manage instruction. These efforts are playing out in a variety of contexts in the US, including in public school districts, non-profits, and other educational networks, and it is shifting how teachers carry out the day-to-day work of instruction. In this comparative case study, we investigate two IFESs in which efforts to improve instruction pushed against historic norms of teacher autonomy. We found that these new systems are not at odds with teacher autonomy, but rather these systems reflect a transition to more interdependent notions of teacher autonomy. 

   more » « less
3. Ride-Sharing Networks with Mixed Autonomy

   https://doi.org/10.23919/ACC.2019.8815333  

   Wei, Qinshuang; Rodriguez, Jorge Alberto; Pedarsani, Ramtin; Coogan, Samuel (July 2019, American Controls Conference)
4. Mixed Autonomy in Ride-Sharing Networks

   https://doi.org/10.1109/TCNS.2020.3016628  

   Wei, Qinshuang; Pedarsani, Ramtin; Coogan, Samuel (January 2021, IEEE Transactions on Control of Network Systems)
5. Residual Policy Learning for Shared Autonomy

   Schaff, Charles; Walter, Matthew R. (July 2020, Proceedings of Robotics: Science and Systems (RSS))

   Shared autonomy provides an effective framework for human-robot collaboration that takes advantage of the complementary strengths of humans and robots to achieve common goals. Many existing approaches to shared autonomy make restrictive assumptions that the goal space, environment dynamics, or human policy are known a priori, or are limited to discrete action spaces, preventing those methods from scaling to complicated real world environments. We propose a model-free, residual policy learning algorithm for shared autonomy that alleviates the need for these assumptions. Our agents are trained to minimally adjust the human’s actions such that a set of goal-agnostic constraints are satisfied. We test our method in two continuous control environments: Lunar Lander, a 2D flight control domain, and a 6-DOF quadrotor reaching task. In experiments with human and surrogate pilots, our method significantly improves task performance without any knowledge of the human’s goal beyond the constraints. These results highlight the ability of model-free deep reinforcement learning to realize assistive agents suited to continuous control settings with little knowledge of user intent. 

   more » « less