However, such models are typically trained on a single modality at a time.

This,however, typically ignores cortical heterogeneity and may smooth out sources ofvariation [11,16]inways which limitinterpretation [7].

Directional features As in previous work [3, 6, 10], we apply a 128-dimensional directional encoding byreplicating(cosθi,sinθi,cosφi,sinφi)by32times torepresent theorientation ofeach single-viewiwith respect to the agent's current orientation, whereθi andφi are the angles of the heading and elevation to that single-view. Replicating the encoding by 32 times does not enrich its information but makes its gradient 32 times larger during back-propagation. We suspect that this benefits the agent to learn about the action-related terms (e.g.

Vision-and-language navigation in the real-world is an important step towards building mobile agents that perceivetheir environments and complete specific tasks following human instructions.

In fact, Gradient Methods (GMs), which can effectively handle BLO problems of largescale,thusgainpopularity.

Inthis paper,weconsider thematching ofmulti-agent multi-armed bandit problem, i.e., while agents prefer arms with higher expected reward, arms also have preferences onagents.