Continual learning agents experience a stream of (related) tasks. The main challenge is that the agent must not forget previous tasks and also adapt to novel tasks in the stream. We are interested in the intersection of two recent continual-learning scenarios. In meta-continual learning, the model is pre-trained using meta-learning to minimize catastrophic forgetting of previous tasks. In continual-meta learning, the aim is to train agents for faster remembering of previous tasks through adaptation.

For readability, we omit OSAKA pre-training. Replay-based methods store representative samples from the past, either in their original form (e.g., rehearsal Most prior-based methods rely on task boundaries. Since non-stationary data distributions breaks the i.i.d assumption for The update is computed from a parametric combination of the gradient of the current and previous task. Despite that, meta-continual learning is actively researched [61, 6]. Bayesian change-point detection scheme to identify whether a task has changed.

Continual learning agents experience a stream of (related) tasks.

Continual learning agents experience a stream of (related) tasks.

Continual learning agents experience a stream of (related) tasks. The main challenge is that the agent must not forget previous tasks and also adapt to novel tasks in the stream. We are interested in the intersection of two recent continual-learning scenarios. In meta-continual learning, the model is pre-trained using meta-learning to minimize catastrophic forgetting of previous tasks. In continual-meta learning, the aim is to train agents for faster remembering of previous tasks through adaptation.

We represent a vehicle dynamics model for autonomous driving near the limits of handling via a multi-layer neural network. Online adaptation is desirable in order to address unseen environments. However, the model needs to adapt to new environments without forgetting previously encountered ones. In this study, we apply Continual-MAML to overcome this difficulty. It enables the model to adapt to the previously encountered environments quickly and efficiently by starting updates from optimized initial parameters. We evaluate the impact of online model adaptation with respect to inference performance and impact on control performance of a model predictive path integral (MPPI) controller using the TRIKart platform. The neural network was pre-trained using driving data collected in our test environment, and experiments for online adaptation were executed on multiple different road conditions not contained in the training data. Empirical results show that the model using Continual-MAML outperforms the fixed model and the model using gradient descent in test set loss and online tracking performance of MPPI.

Learning from non-stationary data remains a great challenge for machine learning. Continual learning addresses this problem in scenarios where the learning agent faces a stream of changing tasks. In these scenarios, the agent is expected to retain its highest performance on previous tasks without revisiting them while adapting well to the new tasks. Two new recent continual-learning scenarios have been proposed. In meta-continual learning, the model is pre-trained to minimize catastrophic forgetting when trained on a sequence of tasks. In continual-meta learning, the goal is faster remembering, i.e., focusing on how quickly the agent recovers performance rather than measuring the agent's performance without any adaptation. Both scenarios have the potential to propel the field forward. Yet in their original formulations, they each have limitations. As a remedy, we propose a more general scenario where an agent must quickly solve (new) out-of-distribution tasks, while also requiring fast remembering. We show that current continual learning, meta learning, meta-continual learning, and continual-meta learning techniques fail in this new scenario. Accordingly, we propose a strong baseline: Continual-MAML, an online extension of the popular MAML algorithm. In our empirical experiments, we show that our method is better suited to the new scenario than the methodologies mentioned above, as well as standard continual learning and meta learning approaches.