In the field of unsupervised skill discovery (USD), a major challenge is limited exploration, primarily due to substantial penalties when skills deviate from their initial trajectories. To enhance exploration, recent methodologies employ auxiliary rewards to maximize the epistemic uncertainty or entropy of states. However, we have identified that the effectiveness of these rewards declines as the environmental complexity rises. Therefore, we present a novel USD algorithm, skill discovery with guidance (DISCO-DANCE), which (1) selects the guide skill that possesses the highest potential to reach unexplored states, (2) guides other skills to follow guide skill, then (3) the guided skills are dispersed to maximize their discriminability in unexplored states. Empirical evaluation demonstrates that DISCO-DANCE outperforms other USD baselines in challenging environments, including two navigation benchmarks and a continuous control benchmark. Qualitative visualizations and code of DISCO-DANCE are available at https://mynsng.github.io/discodance.
In Reinforcement Learning (RL), enhancing sample efficiency is crucial, particularly in scenarios when data acquisition is costly and risky. In principle, off-policy RL algorithms can improve sample efficiency by allowing multiple updates per environment interaction. However, these multiple updates often lead to overfitting, which decreases the network's ability to adapt to new data. We conduct an empirical analysis of this challenge and find that generalizability and plasticity constitute different roles in improving the model's adaptability. In response, we propose a combined usage of Sharpness-Aware Minimization (SAM) and a reset mechanism. SAM seeks wide, smooth minima, improving generalization, while the reset mechanism, through periodic reinitialization of the last few layers, consistently injects plasticity into the model. Through extensive empirical studies, we demonstrate that this combined usage improves sample efficiency and computational cost on the Atari-100k and DeepMind Control Suite benchmarks.
The goal of most materials discovery is to discover materials that are superior to those currently known. Fundamentally, this is close to extrapolation, which is a weak point for most machine learning models that learn the probability distribution of data. Herein, we develop AI-driven combinatorial chemistry, which is a rule-based inverse molecular designer that does not rely on data. Since our model has the potential to generate all possible molecular structures that can be obtained from combinations of molecular fragments, unknown materials with superior properties can be discovered. We theoretically and empirically demonstrate that our model is more suitable for discovering better materials than probability distribution-learning models. In an experiment aimed at discovering molecules that hit seven target properties, our model discovered 1,315 of all target-hitting molecules and 7,629 of five target-hitting molecules out of 100,000 trials, whereas the probability distribution-learning models failed. To illustrate the performance in actual problems, we also demonstrate that our models work well on two practical applications: discovering protein docking materials and HIV inhibitors.
This paper presents a personalized character recommendation system for Multiplayer Online Battle Arena (MOBA) games which are considered as one of the most popular online video game genres around the world. When playing MOBA games, players go through a draft stage, where they alternately select a virtual character to play. When drafting, players select characters by not only considering their character preferences, but also the synergy and competence of their team's character combination. However, the complexity of drafting induces difficulties for beginners to choose the appropriate characters based on the characters of their team while considering their own champion preferences. To alleviate this problem, we propose DraftRec, a novel hierarchical model which recommends characters by considering each player's champion preferences and the interaction between the players. DraftRec consists of two networks: the player network and the match network. The player network captures the individual player's champion preference, and the match network integrates the complex relationship between the players and their respective champions. We train and evaluate our model from a manually collected 280,000 matches of League of Legends and a publicly available 50,000 matches of Dota2. Empirically, our method achieved state-of-the-art performance in character recommendation and match outcome prediction task. Furthermore, a comprehensive user survey confirms that DraftRec provides convincing and satisfying recommendations. Our code and dataset are available at https://github.com/dojeon-ai/DraftRec.
Chemical formula is an artificial language that expresses molecules as text. Neural machines that have learned chemical language can be used as a tool for inverse molecular design. Here, we propose a neural machine that creates molecules that meet some desired conditions based on a deep understanding of chemical language (generative chemical Transformer, GCT). Attention-mechanism in GCT allows a deeper understanding of molecular structures, beyond the limitations of chemical language itself that cause semantic discontinuity, by paying attention to characters sparsely. We investigate the significance of language models to inverse molecular design problems by quantitatively evaluating the quality of generated molecules. GCT generates highly realistic chemical strings that satisfy both a chemical rule and grammars of a language. Molecules parsed from generated strings simultaneously satisfy the multiple target properties and are various for a single condition set. GCT generates de novo molecules, and this is done in a short time that human experts cannot. These advances will contribute to improving the quality of human life by accelerating the process of desired material discovery.