Abstract
While the concept of a conscious machine is intriguing, producing such a machine remains controversial and challenging. Here, we describe how our work on creating a humanoid cognitive robot that learns to perform tasks via imitation learning relates to this issue. Our discussion is divided into three parts. First, we summarize our previous framework for advancing the understanding of the nature of phenomenal consciousness. This framework is based on identifying computational correlates of consciousness. Second, we describe a cognitive robotic system that we recently developed that learns to perform tasks by imitating human-provided demonstrations. This humanoid robot uses cause–effect reasoning to infer a demonstrator’s intentions in performing a task, rather than just imitating the observed actions verbatim. In particular, its cognitive components center on top-down control of a working memory that retains the explanatory interpretations that the robot constructs during learning. Finally, we describe our ongoing work that is focused on converting our robot’s imitation learning cognitive system into purely neurocomputational form, including both its low-level cognitive neuromotor components, its use of working memory, and its causal reasoning mechanisms. Based on our initial results, we argue that the top-down cognitive control of working memory, and in particular its gating mechanisms, is an important potential computational correlate of consciousness in humanoid robots. We conclude that developing high-level neurocognitive control systems for cognitive robots and using them to search for computational correlates of consciousness provides an important approach to advancing our understanding of consciousness, and that it provides a credible and achievable route to ultimately developing a phenomenally conscious machine.
Highlights
In this paper, we use the word “consciousness” to mean phenomenal consciousness unless explicitly indicated otherwise
We replaced traditional low-level motion planning with the DIRECT neural algorithm (Bullock et al, 1993; Gentili et al, 2015), which learns in an unsupervised fashion using exploratory “babbling.” Much of the robotic motion planning done during imitation learning of maintenance tasks requires the use of an inverse kinematics solver that determines a joint trajectory for a given end-effector starting position and target
In terms of the computational explanatory gap (CEG), the issue becomes: can we identify neurocomputational mechanisms that might implement the control of working memory functionality? Elaborating on the hypothesis stated at the beginning of this section, our proposal is that topdown gating like that described above, which determines what is saved and discarded by working memory, furnishes the computational machinery that is used by executive cognitive processes in controlling working memory operations during conscious information processing and is a potential computational correlate of consciousness
Summary
We use the word “consciousness” to mean phenomenal consciousness unless explicitly indicated otherwise. Consciousness is not well understood at present, and many philosophers have questioned whether computational studies or cognitive robots can play a significant role in understanding it Such arguments cannot be refuted at present because there is currently no convincing implementation of instantiated consciousness in a machine, as described in Reggia (2013). As we explain below, both cause–effect reasoning and working memory are widely recognized to be important aspects of conscious human thought, we suggest that exploring how the cognitive and memory mechanisms embodied in our imitation learning robot provide an excellent test of our framework for studying consciousness in machines. We summarize our initial results exploring the feasibility of this idea Based on these results, we argue that the top-down cognitive control of working memory, and its gating mechanisms, is potentially an important computational correlate of consciousness in humanoid robots that merits much further study. We conclude that developing neurocognitive control systems for cognitive robots and using them to search for computational correlates of consciousness provides an important approach to advancing our understanding of consciousness, and that it provides a credible and achievable route to developing a phenomenally conscious machine
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
