We propose Selective Multiple Power Iterations (SMPI), a new algorithm to address the important Tensor PCA problem that consists in recovering a spike $\bf{v_0}^{\otimes k}$ corrupted by a Gaussian noise tensor $\bf{Z} \in (\mathbb{R}^n)^{\otimes k}$ such that $\bf{T}=\sqrt{n} \beta \bf{v_0}^{\otimes k} + \bf{Z}$ where $\beta$ is the signal-to-noise ratio (SNR). SMPI consists in generating a polynomial number of random initializations, performing a polynomial number of symmetrized tensor power iterations on each initialization, then selecting the one that maximizes $\langle \bf{T}, \bf{v}^{\otimes k} \rangle$. Various numerical simulations for $k=3$ in the conventionally considered range $n \leq 1000$ show that the experimental performances of SMPI improve drastically upon existent algorithms and becomes comparable to the theoretical optimal recovery. We show that these unexpected performances are due to a powerful mechanism in which the noise plays a key role for the signal recovery and that takes place at low $\beta$. Furthermore, this mechanism results from five essential features of SMPI that distinguish it from previous algorithms based on power iteration. These remarkable results may have strong impact on both practical and theoretical applications of Tensor PCA. (i) We provide a variant of this algorithm to tackle low-rank CP tensor decomposition. These proposed algorithms also outperforms existent methods even on real data which shows a huge potential impact for practical applications. (ii) We present new theoretical insights on the behavior of SMPI and gradient descent methods for the optimization in high-dimensional non-convex landscapes that are present in various machine learning problems. (iii) We expect that these results may help the discussion concerning the existence of the conjectured statistical-algorithmic gap.
翻译:我们提出选择性多重电流( SMAPI), 这是一种新的算法, 以解决重要的 Tensor CPA 问题, 包括恢复一个峰值 $\bf{v_ 0 ⁇ _ otimes k} 。 由高斯的噪声 ARor $\ bfbb{R ⁇ }\ otime k} (mathbbb{T ⁇ {sqrt{n}\bf{v_ 0 ⁇ xtime k} +\bf ⁇ $, 其中$\beta$ 是信号到音速率比率比率比率比率比率(SNRR) 。 SMPI 生成一个多数值数的随机初始初始初始化信号, 在每个初始初始化时, 并选择一个最小化的电流值( Wefflortial dislorvad), 以实际化的振荡方式, 以虚拟的电流化的电流值为基础( Wev_to time k} 。 在常规的SMIdemodiadeal dal dal dal disal dalation 中, Smardial dismode dismode dism 显示Smodeal demods dismal dislations dismods disl dismods dismods dism lads a lads disms dismods disml lax dismods dismods lads lads lax lads dismod dismods dismods lads lads dismods lads lads lads dismods lads lads lads lads lads lads lads lads lads lads lads lads lads lads lads lads lads lads lads lads lads ladalds las las las las