arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2103.05609cs.LGcs.SYeess.SY
收藏

离散函数基与卷积神经网络

Discrete Function Bases and Convolutional Neural Networks

Centre for Theoretical Neuroscience(理论神经科学中心);University of Waterloo(滑铁卢大学)

Andreas Stöckel

摘要:

我们讨论了“离散函数基”的概念,特别侧重于来自勒让德延迟网络(LDN)的离散基。我们的特点,这些基地的延迟计算任务中的性能,并作为固定的时间卷积神经网络。使用固定时间卷积的网络在概念上很简单,并且在psMNIST等任务中产生最先进的结果。(2)Legendre延迟网络(LDN)可以用来构造一个具有基变换矩阵H的离散函数基。(3)如果q < 300,与LDN基在线卷积比与任意FIR滤波器卷积具有更低的运行时间复杂度。(4)滑动窗口变换存在于一些基(Haar,余弦,傅立叶),每个样本需要O(q)运算和O(N)内存。(5)类似于LDN的LTI系统可以针对许多离散函数基构造; LDN系统在具有有限脉冲响应方面更优越。(6)我们比较离散函数的基础,线性解码延迟信号表示相对于这些基地。结果如图20所示。总的来说,解码错误是相似的。LDN基的误差最大,傅立叶基和余弦基的误差最小。(7)傅立叶和余弦基的特征在于所有延迟的统一解码误差。如果信号可以在傅立叶域中很好地表示,则应使用这些基。(8)神经网络实验表明,固定时间卷积可以优于学习卷积。基的选择并不重要;我们大致观察到与延迟任务相同的性能趋势。(9)如果O(q)的欧拉更新是可行的,并且O(q)的内存需求很重要,那么LDN是小q的正确选择。

英文摘要:

We discuss the notion of "discrete function bases" with a particular focus on the discrete basis derived from the Legendre Delay Network (LDN). We characterize the performance of these bases in a delay computation task, and as fixed temporal convolutions in neural networks. Networks using fixed temporal convolutions are conceptually simple and yield state-of-the-art results in tasks such as psMNIST. Main Results (1) We present a numerically stable algorithm for constructing a matrix of DLOPs L in O(qN) (2) The Legendre Delay Network (LDN) can be used to form a discrete function basis with a basis transformation matrix H in O(qN). (3) If q < 300, convolving with the LDN basis online has a lower run-time complexity than convolving with arbitrary FIR filters. (4) Sliding window transformations exist for some bases (Haar, cosine, Fourier) and require O(q) operations per sample and O(N) memory. (5) LTI systems similar to the LDN can be constructed for many discrete function bases; the LDN system is superior in terms of having a finite impulse response. (6) We compare discrete function bases by linearly decoding delays from signals represented with respect to these bases. Results are depicted in Figure 20. Overall, decoding errors are similar. The LDN basis has the highest and the Fourier and cosine bases have the smallest errors. (7) The Fourier and cosine bases feature a uniform decoding error for all delays. These bases should be used if the signal can be represented well in the Fourier domain. (8) Neural network experiments suggest that fixed temporal convolutions can outperform learned convolutions. The basis choice is not critical; we roughly observe the same performance trends as in the delay task. (9) The LDN is the right choice for small q, if the O(q) Euler update is feasible, and if the low O(q) memory requirement is of importance.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑