This paper introduces Quantum-PEFT that leverages quantum computations for parameter-efficient fine-tuning (PEFT). Unlike other additive PEFT methods, such as low-rank adaptation (LoRA), Quantum-PEFT exploits an underlying full-rank yet surprisingly parameter efficient _quantum unitary parameterization_. With the use of Pauli parameterization, the number of trainable parameters grows only logarithmically with the ambient dimension, as opposed to linearly as in LoRA-based PEFT methods. Quantum-PEFT achieves vanishingly smaller number of trainable parameters than the lowest-rank LoRA as dimensions grow, enhancing parameter efficiency while maintaining a competitive performance. We apply Quantum-PEFT to several transfer learning benchmarks in language and vision, demonstrating significant advantages in parameter efficiency.

In this paper, we study the loss landscape of one-hidden-layer neural networks with ReLU-like activation functions trained with the empirical squared loss using gradient descent (GD). We identify the stationary points of such networks, which significantly slow down loss decrease during training. To capture such points while accounting for the non-differentiability of the loss, the stationary points that we study are directional stationary points, rather than other notions like Clarke stationary points. We show that, if a stationary point does not contain "escape neurons", which are defined with first-order conditions, it must be a local minimum. Moreover, for the scalar-output case, the presence of an escape neuron guarantees that the stationary point is not a local minimum. Our results refine the description of the *saddle-to-saddle* training process starting from infinitesimally small (vanishing) initialization for shallow ReLU-like networks: By precluding the saddle escape types that previous works did not rule out, we advance one step closer to a complete picture of the entire dynamics. Moreover, we are also able to fully discuss how network embedding, which is to instantiate a narrower network with a wider network, reshapes the stationary points.

Modern deep neural networks exhibit heterogeneity across numerous layers of various types such as residuals, multi-head attention, etc., due to varying structures (dimensions, activation functions, etc.), distinct representation characteristics, which impact predictions. We develop a general layer-wise quantization framework with tight variance and code-length bounds, adapting to the heterogeneities over the course of training. We then apply a new layer-wise quantization technique within distributed variational inequalities (VIs), proposing a novel Quantized Optimistic Dual Averaging (QODA) algorithm with adaptive learning rates, which achieves competitive convergence rates for monotone VIs. We empirically show that QODA achieves up to a $150$% speedup over the baselines in end-to-end training time for training Wasserstein GAN on $12+$ GPUs.

Humans are able to segment images effortlessly without supervision using perceptual grouping. Here, we propose a counter-intuitive computational approach to solving unsupervised perceptual grouping and segmentation: that they arise *because* of neural noise, rather than in spite of it. We (1) mathematically demonstrate that under realistic assumptions, neural noise can be used to separate objects from each other; (2) that adding noise in a DNN enables the network to segment images even though it was never trained on any segmentation labels; and (3) that segmenting objects using noise results in segmentation performance that aligns with the perceptual grouping phenomena observed in humans, and is sample-efficient. We introduce the Good Gestalt (GG) datasets --- six datasets designed to specifically test perceptual grouping, and show that our DNN models reproduce many important phenomena in human perception, such as illusory contours, closure, continuity, proximity, and occlusion. Finally, we (4) show that our model improves performance on our GG datasets compared to other tested unsupervised models by $24.9$%. Together, our results suggest a novel unsupervised segmentation method requiring few assumptions, a new explanation for the formation of perceptual grouping, and a novel potential benefit of neural noise.

Quadruped robots possess advantages on different terrains over other types of mobile robots by virtue of their flexible choices of foothold points. It is crucial to integrate terrain perception with motion planning to exploit the potential of quadruped robots. We propose a novel hierarchical terrain-aware control (HTC) framework, which leverages deep reinforcement learning (DRL) for the high-level planner and optimal control for the low-level controller. In general, traditional control methods yield better stability by using an optimization algorithm. In addition, DRL is able to offer more adaptive behavior. Our approach makes full use of the advantages of these two methods and possesses better adaptability and stability in challenging natural environments. Furthermore, the global height map of the terrain serves as visual information for the DRL, which determines the desired footholds for the robot’s leg swings and body postures. Optimal control calculates the torque of the joints on the standing legs to maintain body balance. Our method is tested on various terrains both simulated and real environments. The experimental results show that HTC can effectively enhance the adaptability of the quadruped robot by coordinating body posture.

When it comes to the control system of quadruped robots, deep reinforcement learning (DRL) is considered to be a promising solution. Despite years of development in this field, difficulties remain in guaranteeing the action stability of DRL-based quadruped robots’ locomotion, especially in tough terrain. In this paper, a terrain-aware teacher-student controller integrating a risk assessment network (RAN) is proposed to alleviate this problem. During the training phase, the RAN can evaluate the risk level of historical observation or current state and further guide the update of the policy, thereby assisting the policy in selecting better actions and avoid risky ones. Furthermore, the real-time elevation map is transmitted to the controller as visual information, so that it can perceive the terrain to produce higher performance locomotion. With the aforementioned configuration, we enable a robot to traverse various challenging terrain in simulation and bound or trot stably in the real environment.

A neural-type cell (NTC) is an electronic circuit that mimics the behavior of a neuron. This paper mainly focuses on one type of NTC. It oscillates to produce pulse-coded signals. A complete description of the NTC's oscillation is offered in this paper. The description takes the body effect into consideration, which was never included in previous research work. Moreover, before, it was noticed by researchers that the NTC was able to oscillate only when the input voltage is within a certain range. However, an explanation of the physical essence of the input bounds has yet to be found. To tackle this, this paper provides detailed analysis to the turn-on/off input voltages. Additionally, although there has been a method proposed to calculate the input voltage bounds of an older version of NTC, it was based on numerical approximation and thus lacked universality. In this paper, a new way of calculating the oscillation-supporting input range for the latest version of NTC, the all-MOS NTC, is put forward.