Lifelong learning offers a promising paradigm of building a generalist agent that learns and adapts over its lifespan. Unlike traditional lifelong learning problems in image and text domains, which primarily involve the transfer of declarative knowledge of entities and concepts, lifelong learning in decision-making (LLDM) also necessitates the transfer of procedural knowledge, such as actions and behaviors. To advance research in LLDM, we introduce LIBERO, a novel benchmark of lifelong learning for robot manipulation. Specifically, LIBERO highlights five key research topics in LLDM: 1) how to efficiently transfer declarative knowledge, procedural knowledge, or the mixture of both; 2) how to design effective policy architectures and 3) effective algorithms for LLDM; 4) the robustness of a lifelong learner with respect to task ordering; and 5) the effect of model pretraining for LLDM. We develop an extendible procedural generation pipeline that can in principle generate infinitely many tasks. For benchmarking purpose, we create four task suites (130 tasks in total) that we use to investigate the above-mentioned research topics. To support sample-efficient learning, we provide high-quality human-teleoperated demonstration data for all tasks. Our extensive experiments present several insightful or even unexpected discoveries: sequential finetuning outperforms existing lifelong learning methods in forward transfer, no single visual encoder architecture excels at all types of knowledge transfer, and naive supervised pretraining can hinder agents' performance in the subsequent LLDM. Check the website at https://libero-project.github.io for the code and the datasets.

Creating artificial intelligence (AI) systems capable of demonstrating lifelong learning is a fundamental challenge, and many approaches and metrics have been proposed to analyze algorithmic properties. However, for existing lifelong learning metrics, algorithmic contributions are confounded by task and scenario structure. To mitigate this issue, we introduce an algorithm-agnostic explainable surrogate-modeling approach to estimate latent properties of lifelong learning algorithms. To validate the structure of the surrogate model, we analyze real performance data from a collection of popular lifelong learning approaches and baselines adapted for lifelong classification and lifelong reinforcement learning. Inspired by the way that humans acquire new skills sequentially and improve over time, lifelong or continual learning (Chen & Liu (2018); Silver et al. (2013)) describes the goal of enabling AI systems to learn tasks sequentially over time while improving performance on both previous and future tasks. Lifelong learning has received much attention in the AI community, and many algorithms have been proposed for both supervised (Delange et al. (2021)) and reinforcement learning (Khetarpal et al. (2020)). We include additional review of lifelong learning approaches in Appendix A.2.

In lifelong learning, the tasks (or classes) to be learned arrive sequentially over time in arbitrary order. During training, knowledge from previous tasks can be captured and transferred to subsequent ones to improve sample efficiency. We consider the setting where all target tasks can be represented in the span of a small number of unknown linear or nonlinear features of the input data. We propose a provable lifelong learning algorithm that maintains and refines the internal feature representation. We prove that for any desired accuracy on all tasks, the dimension of the representation remains close to that of the underlying representation. The resulting sample complexity improves significantly on existing bounds. In the setting of linear features, our algorithm is provably efficient and the sample complexity for input dimension $d$, $m$ tasks with $k$ features up to error $\epsilon$ is $\tilde{O}(dk^{1.5}/\epsilon+km/\epsilon)$. We also prove a matching lower bound for any lifelong learning algorithm that uses a single task learner as a black box. Finally, we complement our analysis with an empirical study.

The recent breakthroughs in computer vision have benefited from the availability of large representative datasets (e.g. ImageNet and COCO) for training. Yet, robotic vision poses unique challenges for applying visual algorithms developed from these standard computer vision datasets due to their implicit assumption over non-varying distributions for a fixed set of tasks. Fully retraining models each time a new task becomes available is infeasible due to computational, storage and sometimes privacy issues, while na\"{i}ve incremental strategies have been shown to suffer from catastrophic forgetting. It is crucial for the robots to operate continuously under open-set and detrimental conditions with adaptive visual perceptual systems, where lifelong learning is a fundamental capability. However, very few datasets and benchmarks are available to evaluate and compare emerging techniques. To fill this gap, we provide a new lifelong robotic vision dataset ("OpenLORIS-Object") collected via RGB-D cameras mounted on mobile robots. The dataset embeds the challenges faced by a robot in the real-life application and provides new benchmarks for validating lifelong object recognition algorithms. Moreover, we have provided a testbed of $9$ state-of-the-art lifelong learning algorithms. Each of them involves $48$ tasks with $4$ evaluation metrics over the OpenLORIS-Object dataset. The results demonstrate that the object recognition task in the ever-changing difficulty environments is far from being solved and the bottlenecks are at the forward/backward transfer designs. Our dataset and benchmark are publicly available at \href{https://lifelong-robotic-vision.github.io/dataset/Data_Object-Recognition.html}{\underline{this url}}.

A BSTRACT Current deep neural networks can achieve remarkable performance on a single task. However, when the deep neural network is continually trained on a sequence of tasks, it seems to gradually forget the previous learned knowledge. This phenomenon is referred to as catastrophic forgetting and motivates the field called lifelong learning. The central question in lifelong learning is how to enable deep neural networks to maintain performance on old tasks while learning a new task. In this paper, we introduce a novel and effective lifelong learning algorithm, called MixEd stochastic GrAdient (MEGA), which allows deep neural networks to acquire the ability of retaining performance on old tasks while learning new tasks. Extensive experimental results show that the proposed MEGA algorithm significantly advances the state-of-the-art on all four commonly used lifelong learning benchmarks, reducing the error by up to 18%. 1 I NTRODUCTION A significant step towards artificial general intelligence (AGI) is to enable the learning agent to acquire the ability of remembering past experiences while being trained on a continuum of tasks. Current deep neural networks are capable of achieving remarkable performance on a single task (Goodfellow et al., 2016). However when the network is retrained on a new task, its performance drops drastically on previously trained tasks, a phenomenon which is referred to as catastrophic forgetting (Ratcliff, 1990; Robins, 1995; French, 1999; Kirkpatrick et al., 2017).