English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Humans, even at a very early age, can learn visual concepts and understandgeometry and layout through active interaction with the environment, andgeneralize their compositions to complete tasks described by natural languagesin novel scenes. To mimic such capability, we propose Embodied Concept Learner(ECL) in an interactive 3D environment. Specifically, a robot agent can groundvisual concepts, build semantic maps and plan actions to complete tasks bylearning purely from human demonstrations and language instructions, withoutaccess to ground-truth semantic and depth supervisions from simulations. ECLconsists of: (i) an instruction parser that translates the natural languagesinto executable programs; (ii) an embodied concept learner that grounds visualconcepts based on language descriptions; (iii) a map constructor that estimatesdepth and constructs semantic maps by leveraging the learned concepts; and (iv)a program executor with deterministic policies to execute each program. ECL hasseveral appealing benefits thanks to its modularized design. Firstly, itenables the robotic agent to learn semantics and depth unsupervisedly actinglike babies, e.g., ground concepts through active interaction and perceivedepth by disparities when moving forward. Secondly, ECL is fully transparentand step-by-step interpretable in long-term planning. Thirdly, ECL could bebeneficial for the embodied instruction following (EIF), outperforming previousworks on the ALFRED benchmark when the semantic label is not provided. Also,the learned concept can be reused for other downstream tasks, such as reasoningof object states. Project page: http://ecl.csail.mit.edu/

Embodied Concept Learner: Self-supervised Learning of Concepts and Mapping through Instruction Following