We study computational-statistical gaps for improper learning in sparse linear regression. More specifically, given $n$ samples from a $k$-sparse linear model in dimension $d$, we ask what is the minimum sample complexity to efficiently (in time polynomial in $d$, $k$, and $n$) find a potentially dense estimate for the regression vector that achieves non-trivial prediction error on the $n$ samples. Information-theoretically this can be achieved using $\Theta(k \log (d/k))$ samples. Yet, despite its prominence in the literature, there is no polynomial-time algorithm known to achieve the same guarantees using less than $\Theta(d)$ samples without additional restrictions on the model. Similarly, existing hardness results are either restricted to the proper setting, in which the estimate must be sparse as well, or only apply to specific algorithms. We give evidence that efficient algorithms for this task require at least (roughly) $\Omega(k^2)$ samples. In particular, we show that an improper learning algorithm for sparse linear regression can be used to solve sparse PCA problems (with a negative spike) in their Wishart form, in regimes in which efficient algorithms are widely believed to require at least $\Omega(k^2)$ samples. We complement our reduction with low-degree and statistical query lower bounds for the sparse PCA problems from which we reduce. Our hardness results apply to the (correlated) random design setting in which the covariates are drawn i.i.d. from a mean-zero Gaussian distribution with unknown covariance.
Several photonic microring resonators (MRRs) based analog accelerators have been proposed to accelerate the inference of integer-quantized CNNs with remarkably higher throughput and energy efficiency compared to their electronic counterparts. However, the existing analog photonic accelerators suffer from three shortcomings: (i) severe hampering of wavelength parallelism due to various crosstalk effects, (ii) inflexibility of supporting various dataflows other than the weight-stationary dataflow, and (iii) failure in fully leveraging the ability of photodetectors to perform in-situ accumulations. These shortcomings collectively hamper the performance and energy efficiency of prior accelerators. To tackle these shortcomings, we present a novel Hybrid timE Amplitude aNalog optical Accelerator, called HEANA. HEANA employs hybrid time-amplitude analog optical multipliers (TAOMs) that increase the flexibility of HEANA to support multiple dataflows. A spectrally hitless arrangement of TAOMs significantly reduces the crosstalk effects, thereby increasing the wavelength parallelism in HEANA. Moreover, HEANA employs our invented balanced photo-charge accumulators (BPCAs) that enable buffer-less, in-situ, temporal accumulations to eliminate the need to use reduction networks in HEANA, relieving it from related latency and energy overheads. Our evaluation for the inference of four modern CNNs indicates that HEANA provides improvements of atleast 66x and 84x in frames-per-second (FPS) and FPS/W (energy-efficiency), respectively, for equal-area comparisons, on gmean over two MRR-based analog CNN accelerators from prior work.
We present Aria Everyday Activities (AEA) Dataset, an egocentric multimodal open dataset recorded using Project Aria glasses. AEA contains 143 daily activity sequences recorded by multiple wearers in five geographically diverse indoor locations. Each of the recording contains multimodal sensor data recorded through the Project Aria glasses. In addition, AEA provides machine perception data including high frequency globally aligned 3D trajectories, scene point cloud, per-frame 3D eye gaze vector and time aligned speech transcription. In this paper, we demonstrate a few exemplar research applications enabled by this dataset, including neural scene reconstruction and prompted segmentation. AEA is an open source dataset that can be downloaded from https://www.projectaria.com/datasets/aea/. We are also providing open-source implementations and examples of how to use the dataset in Project Aria Tools https://github.com/facebookresearch/projectaria_tools.
This paper presents a neural vocoder based on a denoising diffusion probabilistic model (DDPM) incorporating explicit periodic signals as auxiliary conditioning signals. Recently, DDPM-based neural vocoders have gained prominence as non-autoregressive models that can generate high-quality waveforms. The neural vocoders based on DDPM have the advantage of training with a simple time-domain loss. In practical applications, such as singing voice synthesis, there is a demand for neural vocoders to generate high-fidelity speech waveforms with flexible pitch control. However, conventional DDPM-based neural vocoders struggle to generate speech waveforms under such conditions. Our proposed model aims to accurately capture the periodic structure of speech waveforms by incorporating explicit periodic signals. Experimental results show that our model improves sound quality and provides better pitch control than conventional DDPM-based neural vocoders.
Decision-directed channel estimation (DDCE) is one kind of blind channel estimation method that tracks the channel blindly by an iterative algorithm without relying on the pilots, which can increase the utilization of wireless resource. However, one major problem of DDCE is the performance degradation caused by error accumulation during the tracking process. In this paper, we propose an reliable DDCE (RDDCE) scheme for an OFDM-based communication system in the time-varying deep fading environment. By combining the conventional DDCE and discrete Fourier transform (DFT) channel estimation method, the proposed RDDCE scheme selects the reliable estimated channels on the subcarriers which are less affected by deep fading, and then estimates the channel based on the selected subcarriers by an extended DFT channel estimation where the indices of selected subcarriers are not distributed evenly. Simulation results show that RRDCE can alleviate the performance degradation effectively, track the channel with high accuracy on a long time scale, and has good performance under time-varying and noisy channel conditions.
Large-scale recordings of neural activity are providing new opportunities to study neural population dynamics. A powerful method for analyzing such high-dimensional measurements is to deploy an algorithm to learn the low-dimensional latent dynamics. LFADS (Latent Factor Analysis via Dynamical Systems) is a deep learning method for inferring latent dynamics from high-dimensional neural spiking data recorded simultaneously in single trials. This method has shown a remarkable performance in modeling complex brain signals with an average inference latency in milliseconds. As our capacity of simultaneously recording many neurons is increasing exponentially, it is becoming crucial to build capacity for deploying low-latency inference of the computing algorithms. To improve the real-time processing ability of LFADS, we introduce an efficient implementation of the LFADS models onto Field Programmable Gate Arrays (FPGA). Our implementation shows an inference latency of 41.97 $\mu$s for processing the data in a single trial on a Xilinx U55C.
The reconstruction of 3D microstructures from 2D slices is considered to hold significant value in predicting the spatial structure and physical properties of materials.The dimensional extension from 2D to 3D is viewed as a highly challenging inverse problem from the current technological perspective.Recently,methods based on generative adversarial networks have garnered widespread attention.However,they are still hampered by numerous limitations,including oversimplified models,a requirement for a substantial number of training samples,and difficulties in achieving model convergence during training.In light of this,a novel generative model that integrates the multiscale properties of U-net with and the generative capabilities of GAN has been proposed.Based on this,the innovative construction of a multi-scale channel aggregation module,a multi-scale hierarchical feature aggregation module and a convolutional block attention mechanism can better capture the properties of the material microstructure and extract the image information.The model's accuracy is further improved by combining the image regularization loss with the Wasserstein distance loss.In addition,this study utilizes the anisotropy index to accurately distinguish the nature of the image,which can clearly determine the isotropy and anisotropy of the image.It is also the first time that the generation quality of material samples from different domains is evaluated and the performance of the model itself is compared.The experimental results demonstrate that the present model not only shows a very high similarity between the generated 3D structures and real samples but is also highly consistent with real data in terms of statistical data analysis.
Most efforts in Computer Vision focus on natural images or artwork, which differ significantly both in size and contents from the kind of data biomedical image processing deals with. Thus, Transfer Learning models often prove themselves suboptimal for these tasks, even after manual finetuning. The development of architectures from scratch is oftentimes unfeasible due to the vastness of the hyperparameter space and a shortage of time, computational resources and Deep Learning experts in most biomedical research laboratories. An alternative to manually defining the models is the use of Neuroevolution, which employs metaheuristic techniques to optimize Deep Learning architectures. However, many algorithms proposed in the neuroevolutive literature are either too unreliable or limited to a small, predefined region of the hyperparameter space. To overcome these shortcomings, we propose the Chimera Algorithm, a novel, hybrid neuroevolutive algorithm that integrates the Artificial Bee Colony Algorithm with Evolutionary Computation tools to generate models from scratch, as well as to refine a given previous architecture to better fit the task at hand. The Chimera Algorithm has been validated with two datasets of natural and medical images, producing models that surpassed the performance of those coming from Transfer Learning.
Reference features from a template or historical frames are crucial for visual object tracking. Prior works utilize all features from a fixed template or memory for visual object tracking. However, due to the dynamic nature of videos, the required reference historical information for different search regions at different time steps is also inconsistent. Therefore, using all features in the template and memory can lead to redundancy and impair tracking performance. To alleviate this issue, we propose a novel tracking paradigm, consisting of a relevance attention mechanism and a global representation memory, which can adaptively assist the search region in selecting the most relevant historical information from reference features. Specifically, the proposed relevance attention mechanism in this work differs from previous approaches in that it can dynamically choose and build the optimal global representation memory for the current frame by accessing cross-frame information globally. Moreover, it can flexibly read the relevant historical information from the constructed memory to reduce redundancy and counteract the negative effects of harmful information. Extensive experiments validate the effectiveness of the proposed method, achieving competitive performance on five challenging datasets with 71 FPS.
Forecasting timeseries based upon measured data is needed in a wide range of applications and has been the subject of extensive research. A particularly challenging task is the forecasting of timeseries generated by chaotic dynamics. In recent years reservoir computing has been shown to be an effective method of forecasting chaotic dynamics and reconstructing chaotic attractors from data. In this work strides are made toward smaller and lower complexity reservoirs with the goal of improved hardware implementability and more reliable production of adequate surrogate models. We show that a reservoir of uncoupled nodes more reliably produces long term timeseries predictions than complex reservoir topologies. We then link the improved attractor reconstruction of the uncoupled reservoir with smaller spectral radii of the resulting surrogate systems. These results indicate that, the node degree plays an important role in determining whether the desired dynamics will be stable in the autonomous surrogate system which is attained via closed-loop operation of the trained reservoir. In terms of hardware implementability, uncoupled nodes would allow for greater freedom in the hardware architecture because no complex coupling setups are needed and because, for uncoupled nodes, the system response is equivalent for space and time multiplexing.