Learning from Trajectories via Subgoal Discovery

Authors: Sujoy Paul, Jeroen Vanbaar, Amit Roy-Chowdhury

NeurIPS 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We perform experiments on three goal-oriented tasks on Mu Jo Co [15] with sparse terminal-only reward, which state-of-the-art RL, IL or their combinations are not able to solve.
Researcher Affiliation Collaboration 1University of California-Riverside 2Mitsubishi Electric Research Laboratories (MERL)
Pseudocode Yes Algorithm 1 Learning Sub-Goal Prediction
Open Source Code No The paper does not include an unambiguous statement about releasing code or a link to a code repository for the described methodology.
Open Datasets Yes We perform experiments on three challenging environments as shown in Fig. 2. First is Ballin-Maze Game (Bi MGame) introduced in [43]... The second environment is Ant Target which involves the Ant [44]... The third environment, Ant Maze, uses the same Ant, but in a U-shaped maze used in [35].
Dataset Splits No The paper does not provide specific dataset split information (exact percentages, sample counts, citations to predefined splits, or detailed splitting methodology) for training, validation, or testing.
Hardware Specification No The paper does not provide specific hardware details (exact GPU/CPU models, processor types with speeds, memory amounts, or detailed computer specifications) used for running its experiments.
Software Dependencies No The paper does not provide specific ancillary software details, such as library or solver names with version numbers, needed to replicate the experiment.
Experiment Setup Yes Details about the network architectures we use for πθ, πφ and fψ(s) can be found in the supplementary material.