爱可可AI前沿推介(3.7)

LG - 机器学习 CV - 计算机视觉 CL - 计算与语言 AS - 音频与语音 RO - 机器人

转自爱可可爱生活

1、[LG] Continuous-Time Meta-Learning with Forward Mode Differentiation

T Deleu, D Kanaa, L Feng, G Kerg, Y Bengio, G Lajoie, P Bacon

[Mila – Université de Montréal]

基于正向模式微分的连续时间元学习。从具有无限小梯度步骤的基于梯度的元学习方法中得到启发，本文提出连续时间元学习(COMLN)，一种元学习算法，其自适应遵循梯度矢量场的动态。输入的表示被元学习，得到一个特定任务的线性分类器作为常微分方程(ODE)的解。将学习过程视为ODE，提供一个明显的优势，即轨迹的长度现在是连续的，而不是固定的和离散的梯度步骤的数量。因此，除了像基于梯度的元学习的标准做法那样学习初始条件外，还可以优化用随机梯度下降解决新任务所需的自适应量。重要的是，为了计算外层循环更新所需的精确元梯度，设计了一种基于前向模式微分的高效算法，其内存要求不随学习轨迹的长度而变化，因此允许在恒定内存中进行更长时间的自适应。本文为COMLN的稳定性提供了分析性的保证，以经验的方式展示了它在运行时间和内存使用方面的效率，并在一系列的sh图爱样本像分类问题上说明了其有效性。

Drawing inspiration from gradient-based meta-learning methods with infinitely small gradient steps, we introduce Continuous-Time Meta-Learning (COMLN), a meta-learning algorithm where adaptation follows the dynamics of a gradient vector field. Specifically, representations of the inputs are meta-learned such that a task-specific linear classifier is obtained as a solution of an ordinary differential equation (ODE). Treating the learning process as an ODE offers the notable advantage that the length of the trajectory is now continuous, as opposed to a fixed and discrete number of gradient steps. As a consequence, we can optimize the amount of adaptation necessary to solve a new task using stochastic gradient descent, in addition to learning the initial conditions as is standard practice in gradient-based meta-learning. Importantly, in order to compute the exact meta-gradients required for the outer-loop updates, we devise an efficient algorithm based on forward mode differentiation, whose memory requirements do not scale with the length of the learning trajectory, thus allowing longer adaptation in constant memory. We provide analytical guarantees for the stability of COMLN, we show empirically its efficiency in terms of runtime and memory usage, and we illustrate its effectiveness on a range of few-shot image classification problems.

2、[LG] Learning Robust Real-Time Cultural Transmission without Human Data

C G I Team, A Bhoopchand, B Brownfield, A Collister, A D Lago, A Edwards, R Everett, A Frechette, Y G Oliveira, E Hughes, K W. Mathewson, P Mendolicchio, J Pawar, M Pislar, A Platonov, E Senter, S Singh, A Zacherl, L M. Zhang

[DeepMind]

无需人工数据的鲁棒实时文化传播学习。文化传播是一种领域性的社会技能，让智能体以高保真度和高记忆力实时地从对方那里获得和使用信息。对于人类，它是为累积性文化进化提供动力的继承过程，使人类的技能、工具和知识跨代扩展。本文提供了一种在人工智能智能体中产生零样本、高召回率文化传播的方法。所设计的智能体在不使用任何预先收集的人工数据的情况下，成功地从人类那里获得实时的文化传播。本文确定了一套足以产生文化传播的令人惊讶的简单成分，并开发了一种严格评估的评价方法。为文化进化作为发展人工通用智能的一种算法铺平了道路。

Cultural transmission is the domain-general social skill that allows agents to acquire and use information from each other in real-time with high fidelity and recall. In humans, it is the inheritance process that powers cumulative cultural evolution, expanding our skills, tools and knowledge across generations. We provide a method for generating zero-shot, high recall cultural transmission in artificially intelligent agents. Our agents succeed at real-time cultural transmission from humans in novel contexts without using any pre-collected human data. We identify a surprisingly simple set of ingredients sufficient for generating cultural transmission and develop an evaluation methodology for rigorously assessing it. This paves the way for cultural evolution as an algorithm for developing artificial general intelligence.

3、[CL] A Systematic Evaluation of Large Language Models of Code

F F. Xu, U Alon, G Neubig, V J. Hellendoorn

[CMU]

大型程序语言模型的系统评估。最近，程序的大型语言模型(LM)在代码补完和用自然语言描述合成代码方面显示出巨大的前景。然而，目前最先进的代码语言模型并没有公开提供，这就给他们的模型和数据设计决策留下了许多疑问。本文旨在通过对现有最大的模型进行系统评估，填补其中的一些空白。Codex、GPT-J、GPT-Neo、GPT-NeoX-20B和CodeParrot，跨越各种编程语言。尽管Codex本身不是开源的，但现有的开源模型在一些编程语言中确实取得了接近的结果，尽管主要针对的是自然语言建模。进一步确定了一个重要的缺失，那就是专门在多语言的代码语料库中训练的大型开源模型。本文发布了一个新的模型，PolyCoder，具有27B的参数，基于GPT-2架构，在一台机器上对249GB的跨12种编程语言的代码进行了训练。在C编程语言中，PolyCoder优于包括Codex在内的所有模型。所训练的模型是开源的，这使得未来在这个领域的研究和应用成为可能。

Large language models (LMs) of code have recently shown tremendous promise in completing code and synthesizing code from natural language descriptions. However, the current state-of-the-art code LMs (e.g., Codex (Chen et al., 2021)) are not publicly available, leaving many questions about their model and data design decisions. We aim to fill in some of these blanks through a systematic evaluation of the largest existing models: Codex, GPT-J, GPT-Neo, GPT-NeoX-20B, and CodeParrot, across various programming languages. Although Codex itself is not open-source, we find that existing open-source models do achieve close results in some programming languages, although targeted mainly for natural language modeling. We further identify an important missing piece in the form of a large open-source model trained exclusively on a multi-lingual corpus of code. We release a new model, PolyCoder, with 2.7B parameters based on the GPT-2 architecture, which was trained on 249GB of code across 12 programming languages on a single machine. In the C programming language, PolyCoder outperforms all models including Codex. Our trained models are open-source and publicly available at this https URL, which enables future research and application in this area.

4、[RO] Collision-free Path Planning on Arbitrary Optimization Criteria in the Latent Space through cGANs

T Ando, H Iino, H Mori, R Torishima, K Takahashi, S Yamaguchi, D Okanohara, T Ogata

[Waseda Univ & SoftBank Corp & Preferred Networks]

基于cGAN的潜空间任意优化标准无碰撞路径规划。本文提出一种通过条件生成对抗网络(cGAN)进行无碰撞路径规划的新方法，当障碍物图被给定为条件时，其潜空间只映射到机器人关节空间的无碰撞区域。在操纵机器人手臂时，出于安全考虑，有必要生成一条避免与机器人本身或周围环境接触的轨迹，而且生成适合各自目的的多个任意轨迹也很方便。在所提出的方法中，通过在这个潜空间中用任意线段连接起点和目标，可以生成各种避免障碍物的轨迹。所提出方法只是提供了这个无碰撞的潜空间，任意规划器，使用任意优化条件，都可以用来生成最合适的飞行路径。用一个模拟的和实际的UR5e 6-DoF机械臂成功地验证了该方法。实验表明，所提出方法根据不同的优化条件，可以生成不同的轨迹。

We propose a new method for collision-free path planning by Conditional Generative Adversarial Networks (cGANs) by mapping its latent space to only the collision-free areas of the robot joint space when an obstacle map is given as a condition. When manipulating a robot arm, it is necessary to generate a trajectory that avoids contact with the robot itself or the surrounding environment for safety reasons, and it is convenient to generate multiple arbitrary trajectories appropriate for respective purposes. In the proposed method, various trajectories to avoid obstacles can be generated by connecting the start and goal with arbitrary line segments in this latent space. Our method simply provides this collision-free latent space after which any planner, using any optimization conditions, can be used to generate the most suitable paths on the fly. We successfully verified this method with a simulated and actual UR5e 6-DoF robotic arm. We confirmed that different trajectories can be generated according to different optimization conditions.

5、[CV] D²ETR: Decoder-Only DETR with Computationally Efficient Cross-Scale Attention

J Lin, X Mao, Y Chen, L Xu, Y He, H Xue

[Nanjing University & Alibaba Group]

D²ETR：计算高效跨尺度注意力纯解码器DETR。DETR是第一个完全端到端的检测器，无需后处理就能预测出最后一组预测结果。然而，它存在着性能低、收敛慢等问题。一系列的工作旨在以不同的方式解决这些问题，但由于复杂的编码器-解码器结构，计算成本还是很高。为缓解这个问题，本文提出一种仅有解码器的检测器D²ETR。在没有编码器的情况下，解码器直接关注由Transformer骨干网产生的精细融合特征图，并有一个新的计算高效的跨尺度注意力模块。在对COCO基准的评估中，D²ETR表现出较低的计算复杂性和较高的检测精度，超过了DETR及其变体。

DETR is the first fully end-to-end detector that predicts a final set of predictions without post-processing. However, it suffers from problems such as low performance and slow convergence. A series of works aim to tackle these issues in different ways, but the computational cost is yet expensive due to the sophisticated encoder-decoder architecture. To alleviate this issue, we propose a decoder-only detector called D²ETR. In the absence of encoder, the decoder directly attends to the fine-fused feature maps generated by the Transformer backbone with a novel computationally efficient cross-scale attention module. D²ETR demonstrates low computational complexity and high detection accuracy in evaluations on the COCO benchmark, outperforming DETR and its variants.