However, in domains where precise and succinct expert state information is available, agents trained onsuchexpert state features usually outperform agents trained onrichobservations.

Existing methods can be broadly divided into two categories:importance sampling(IS) based and model-based [58].

In recent years, deep generative models have been shown to'imagine' convincing

Long-horizon decision-making with sparse rewards and continuous states and actions remains a fundamental challenge in AI and robotics. Task and motion planning (TAMP) is a model-based framework that addresses this challenge by planning hierarchically with abstract actions (options). These options are manually defined, limiting the agent to behaviors that we as human engineers know how to program (pick, place, move). In this work, we propose Shortcut Learning for Abstract Planning (SLAP), a method that leverages existing TAMP options to automatically discover new ones. Our key idea is to use model-free reinforcement learning (RL) to learn shortcuts in the abstract planning graph induced by the existing options in TAMP. Without any additional assumptions or inputs, shortcut learning leads to shorter solutions than pure planning, and higher task success rates than flat and hierarchical RL. Qualitatively, SLAP discovers dynamic physical improvisations (e.g., slap, wiggle, wipe) that differ significantly from the manually-defined ones. In experiments in four simulated robotic environments, we show that SLAP solves and generalizes to a wide range of tasks, reducing overall plan lengths by over 50% and consistently outperforming planning and RL baselines.

In particular, DNN-specific overestimation is extremely unpredictable due to many factors such as the dimension of system states, the complexity of environment dynamics, and the size, weight, and activation function of a neural network.

So far, these methods have only been used to discover policies over simple, finite action spaces.