Parsa Nooralinejad

Soroush Abbasi Koohpayegani, KL Navaneet, Parsa Nooralinejad et al.

Fine-tuning Large Language Models (LLMs) and storing them for each downstream task or domain is impractical because of the massive model size (e.g., 350GB in GPT-3). Current literature, such as LoRA, showcases the potential of low-rank modifications to the original weights of an LLM, enabling efficient adaptation and storage for task-specific models. These methods can reduce the number of parameters needed to fine-tune an LLM by several orders of magnitude. Yet, these methods face two primary limitations: (1) the parameter count is lower-bounded by the rank one decomposition, and (2) the extent of reduction is heavily influenced by both the model architecture and the chosen rank. We introduce NOLA, which overcomes the rank one lower bound present in LoRA. It achieves this by re-parameterizing the low-rank matrices in LoRA using linear combinations of randomly generated matrices (basis) and optimizing the linear mixture coefficients only. This approach allows us to decouple the number of trainable parameters from both the choice of rank and the network architecture. We present adaptation results using GPT-2, LLaMA-2, and ViT in natural language and computer vision tasks. NOLA performs as well as LoRA models with much fewer number of parameters compared to LoRA with rank one, the best compression LoRA can archive. Particularly, on LLaMA-2 70B, our method is almost 20 times more compact than the most compressed LoRA without degradation in accuracy. Our code is available here: https://github.com/UCDvision/NOLA

Parsa Nooralinejad, Ali Abbasi, Soroush Abbasi Koohpayegani et al.

We demonstrate that a deep model can be reparametrized as a linear combination of several randomly initialized and frozen deep models in the weight space. During training, we seek local minima that reside within the subspace spanned by these random models (i.e., `basis' networks). Our framework, PRANC, enables significant compaction of a deep model. The model can be reconstructed using a single scalar `seed,' employed to generate the pseudo-random `basis' networks, together with the learned linear mixture coefficients. In practical applications, PRANC addresses the challenge of efficiently storing and communicating deep models, a common bottleneck in several scenarios, including multi-agent learning, continual learners, federated systems, and edge devices, among others. In this study, we employ PRANC to condense image classification models and compress images by compacting their associated implicit neural networks. PRANC outperforms baselines with a large margin on image classification when compressing a deep model almost $100$ times. Moreover, we show that PRANC enables memory-efficient inference by generating layer-wise weights on the fly. The source code of PRANC is here: \url{https://github.com/UCDvision/PRANC}

Ali Abbasi, Parsa Nooralinejad, Vladimir Braverman et al.

Continual/lifelong learning from a non-stationary input data stream is a cornerstone of intelligence. Despite their phenomenal performance in a wide variety of applications, deep neural networks are prone to forgetting their previously learned information upon learning new ones. This phenomenon is called "catastrophic forgetting" and is deeply rooted in the stability-plasticity dilemma. Overcoming catastrophic forgetting in deep neural networks has become an active field of research in recent years. In particular, gradient projection-based methods have recently shown exceptional performance at overcoming catastrophic forgetting. This paper proposes two biologically-inspired mechanisms based on sparsity and heterogeneous dropout that significantly increase a continual learner's performance over a long sequence of tasks. Our proposed approach builds on the Gradient Projection Memory (GPM) framework. We leverage k-winner activations in each layer of a neural network to enforce layer-wise sparse activations for each task, together with a between-task heterogeneous dropout that encourages the network to use non-overlapping activation patterns between different tasks. In addition, we introduce two new benchmarks for continual learning under distributional shift, namely Continual Swiss Roll and ImageNet SuperDog-40. Lastly, we provide an in-depth analysis of our proposed method and demonstrate a significant performance boost on various benchmark continual learning problems.

Ali Abbasi, Parsa Nooralinejad, Hamed Pirsiavash et al.

Continual learning has gained substantial attention within the deep learning community, offering promising solutions to the challenging problem of sequential learning. Yet, a largely unexplored facet of this paradigm is its susceptibility to adversarial attacks, especially with the aim of inducing forgetting. In this paper, we introduce "BrainWash," a novel data poisoning method tailored to impose forgetting on a continual learner. By adding the BrainWash noise to a variety of baselines, we demonstrate how a trained continual learner can be induced to forget its previously learned tasks catastrophically, even when using these continual learning baselines. An important feature of our approach is that the attacker requires no access to previous tasks' data and is armed merely with the model's current parameters and the data belonging to the most recent task. Our extensive experiments highlight the efficacy of BrainWash, showcasing degradation in performance across various regularization-based continual learning methods.

Chayne Thrash, Ali Abbasi, Reed Andreas et al.

The outstanding performance of large foundational models across diverse tasks, from computer vision to speech and natural language processing, has significantly increased their demand. However, storing and transmitting these models poses significant challenges due to their massive size (e.g., 750GB for Llama 3.1 405B). Recent literature has focused on compressing the original weights or reducing the number of parameters required for fine-tuning these models. These compression methods generally constrain the parameter space, for example, through low-rank reparametrization (e.g., LoRA), pruning, or quantization (e.g., QLoRA) during or after the model training. In this paper, we present a novel model compression method, which we term Manifold-Constrained Neural Compression (MCNC). This method constrains the parameter space to low-dimensional pre-defined and frozen nonlinear manifolds, which effectively cover this space. Given the prevalence of good solutions in over-parameterized deep neural networks, we show that by constraining the parameter space to our proposed manifold, we can identify high-quality solutions while achieving unprecedented compression rates across a wide variety of tasks and architectures. Through extensive experiments in computer vision and natural language processing tasks, we demonstrate that our method significantly outperforms state-of-the-art baselines in terms of compression, accuracy, and/or model reconstruction time. Our code is publicly available at https://github.com/mint-vu/MCNC.

Kossar Pourahmadi, Parsa Nooralinejad, Hamed Pirsiavash

Active learning focuses on choosing a subset of unlabeled data to be labeled. However, most such methods assume that a large subset of the data can be annotated. We are interested in low-budget active learning where only a small subset (e.g., 0.2% of ImageNet) can be annotated. Instead of proposing a new query strategy to iteratively sample batches of unlabeled data given an initial pool, we learn rich features by an off-the-shelf self-supervised learning method only once, and then study the effectiveness of different sampling strategies given a low labeling budget on a variety of datasets including ImageNet. We show that although the state-of-the-art active learning methods work well given a large labeling budget, a simple K-means clustering algorithm can outperform them on low budgets. We believe this method can be used as a simple baseline for low-budget active learning on image classification. Code is available at: https://github.com/UCDvision/low-budget-al

Reza Hojabr, Kamyar Givaki, Kossar Pourahmadi et al.

Emerging intelligent embedded devices rely on Deep Neural Networks (DNNs) to be able to interact with the real-world environment. This interaction comes with the ability to retrain DNNs, since environmental conditions change continuously in time. Stochastic Gradient Descent (SGD) is a widely used algorithm to train DNNs by optimizing the parameters over the training data iteratively. In this work, first we present a novel approach to add the training ability to a baseline DNN accelerator (inference only) by splitting the SGD algorithm into simple computational elements. Then, based on this heuristic approach we propose TaxoNN, a light-weight accelerator for DNN training. TaxoNN can easily tune the DNN weights by reusing the hardware resources used in the inference process using a time-multiplexing approach and low-bitwidth units. Our experimental results show that TaxoNN delivers, on average, 0.97% higher misclassification rate compared to a full-precision implementation. Moreover, TaxoNN provides 2.1$\times$ power saving and 1.65$\times$ area reduction over the state-of-the-art DNN training accelerator.