Zehua Pei, Hui-Ling Zhen, Lancheng Zou, Xianzhi Yu, Wulong Liu, Sinno Jialin Pan, Mingxuan Yuan, Bei Yu. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

KV cache quantization can improve Large Language Models (LLMs) inference throughput and latency in long contexts and large batch-size scenarios while preserving LLMs effectiveness. However, current methods have three unsolved issues: overlooking layer-wise sensitivity to KV cache quantization, high overhead of online fine-grained decision-making, and low flexibility to different LLMs and constraints. Therefore, we theoretically analyze the inherent correlation of layer-wise transformer attention patterns to KV cache quantization errors and study why key cache is generally more important than value cache for quantization error reduction. We further propose a simple yet effective framework KVTuner to adaptively search for the optimal hardware-friendly layer-wise KV quantization precision pairs for coarse-grained KV cache with multi-objective optimization and directly utilize the offline searched configurations during online inference. To reduce the computational cost of offline calibration, we utilize the intra-layer KV precision pair pruning and inter-layer clustering to reduce the search space. Experimental results show that we can achieve nearly lossless 3.25-bit mixed precision KV cache quantization for LLMs like Llama-3.1-8B-Instruct and 4.0-bit for sensitive models like Qwen2.5-7B-Instruct on mathematical reasoning tasks. The maximum inference throughput can be improved by 21.25% compared with KIVI-KV8 quantization over various context lengths. Our code and searched configurations are available at https://github.com/cmd2001/KVTuner.

Recently, quantization has been widely used for the compression and acceleration of large language models (LLMs). Due to the outliers in LLMs, it is crucial to flatten weights and activations to minimize quantization error with equally spaced quantization points. Prior research explores various pre-quantization transformations to suppress outliers, such as per-channel scaling and Hadamard transformation. However, we observe that these transformed weights and activations can still exhibit steep and dispersed distributions. In this paper, we propose FlatQuant (Fast and Learnable Affine Transformation), a new post-training quantization approach that enhances the flatness of weights and activations. Our approach identifies optimal affine transformations for each linear layer, calibrated in hours via a lightweight objective. To reduce runtime overhead of affine transformation, we apply Kronecker product with two lightweight matrices, and fuse all operations in FlatQuant into a single kernel. Extensive experiments demonstrate that FlatQuant establishes a new state-of-the-art benchmark for quantization. For example, it achieves less than 1\% accuracy drop for W4A4 quantization on the LLaMA-3-70B model, surpassing SpinQuant by 7.5\%. Additionally, it provides up to 2.3x prefill speedup and 1.7x decoding speedup compared to the FP16 model. Code is available at: https://github.com/ruikangliu/FlatQuant.

Current inference systems for Mixture-of-Experts (MoE) models primarily employ static parallelization strategies. However, these static approaches cannot consistently achieve optimal performance across different inference scenarios, as they lack the flexibility to adapt to varying computational requirements. In this work, we propose HAP (Hybrid Adaptive Parallelism), a novel method that dynamically selects hybrid parallel strategies to enhance MoE inference efficiency. The fundamental innovation of HAP lies in hierarchically decomposing MoE architectures into two distinct computational modules: the Attention module and the Expert module, each augmented with a specialized inference latency simulation model. This decomposition promotes the construction of a comprehensive search space for seeking model parallel strategies. By leveraging Integer Linear Programming (ILP), HAP could solve the optimal hybrid parallel configurations to maximize inference efficiency under varying computational constraints. Our experiments demonstrate that HAP consistently determines parallel configurations that achieve comparable or superior performance to the TP strategy prevalent in mainstream inference systems. Compared to the TP-based inference, HAP-based inference achieves speedups of$1.68 \times, 1.77 \times$, and$1.57 \times$on A100, A6000, and V100 GPU platforms, respectively. Furthermore, HAP showcases remarkable generalization capability, maintaining performance effectiveness across diverse MoE model configurations, including Mixtral and Qwen series models.

With the development of large language models (LLMs), efficient inference through Key-Value (KV) cache compression has attracted considerable attention, especially for long-context generation. To compress the KV cache, recent methods identify critical KV tokens through static modeling of attention scores. However, these methods often struggle to accurately determine critical tokens as they neglect the *temporal patterns* in attention scores, resulting in a noticeable degradation in LLM performance. To address this challenge, we propose **AttentionPredictor**, which is the **first learning-based method to directly predict attention patterns for KV cache compression and critical token identification**. Specifically, AttentionPredictor learns a lightweight, unified convolution model to dynamically capture spatiotemporal patterns and predict the next-token attention scores. An appealing feature of AttentionPredictor is that it accurately predicts the attention score and shares the unified prediction model, which consumes negligible memory, among all transformer layers. Moreover, we propose a cross-token critical cache prefetching framework that hides the token estimation time overhead to accelerate the decoding stage. By retaining most of the attention information, AttentionPredictor achieves **13$\times$** KV cache compression and **5.6$\times$** speedup in a cache offloading scenario with comparable LLM performance, significantly outperforming the state-of-the-arts. The code is available at https://github.com/MIRALab-USTC/LLM-AttentionPredictor.

In the fast-evolving artificial intelligence era, the intersection of natural language processing and financial analysis has attracted significant attention, primarily due to its potential to provide valuable insights into financial market behavior. Sentiment analysis of financial news articles is a crucial aspect of this intersection, providing cues about market sentiment that may affect stock price dynamics. Traditional sentiment analysis methods often rely on rules or machine learning algorithms trained on labeled datasets, but these methods face challenges in capturing the context within the text. This paper proposes a framework that incorporates prompt engineering strategies, including a novel Domain Knowledge Chain-of-Thought (DK-CoT) strategy, integrating domain-specific financial knowledge with chain-of-thought reasoning, designed to leverage and enhance the performance of large language models (LLMs) in financial news sentiment analysis. DK-CoT has been compared with various prompt engineering techniques, including zero-shot, few-shot, and chain-of-thought, as well as other benchmark models like BERT and RoBERTa. Through comprehensive experiments and evaluations, we introduce the weighted F1 score as a more practical metric, emphasizing the disproportionate impact of negative news on financial markets, which better reflects real-world financial dynamics, as negative sentiments often lead to more significant market reactions than positive or neutral sentiments. Experimental results have shown that DK-CoT adopted in an LLM called GLM is effective in improving the performance and reliability of financial news sentiment analysis. Our findings provide insights into optimal prompt designs and highlight the importance of incorporating financial knowledge to uplift LLM performance while reducing the need for extensive computational resources and fine-tuning.

The problem of transistor sizing is challenging due to the large design space and complex performance trade-offs. Conventional black-box optimization methods, such as Bayesian optimization, cannot leverage past experience. In this paper, we propose a novel state representation for analog circuits to capture the heterogeneity of connections between different components. Component types are encoded through an embedding lookup table, which enables learning transferable knowledge across circuits. Experiments on various designs demonstrate that the agent with transfer learning can reduce runtime significantly in both fine-tuning and zero-shot transfer settings compared to current SOTA baselines. Notably, when learning from scratch, our agent achieves at least 21% higher Figures of Merit (FoM) compared to the SOTA method.

The empirical success of distributional reinforcement learning (RL) highly relies on the choice of distribution divergence equipped with an appropriate distribution representation. In this paper, we propose \textit{Sinkhorn distributional RL (SinkhornDRL)}, which leverages Sinkhorn divergence—a regularized Wasserstein loss—to minimize the difference between current and target Bellman return distributions. Theoretically, we prove the contraction properties of SinkhornDRL, aligning with the interpolation nature of Sinkhorn divergence between Wasserstein distance and Maximum Mean Discrepancy (MMD). The introduced SinkhornDRL enriches the family of distributional RL algorithms, contributing to interpreting the algorithm behaviors compared with existing approaches by our investigation into their relationships. Empirically, we show that SinkhornDRL consistently outperforms or matches existing algorithms on the Atari games suite and particularly stands out in the multi-dimensional reward setting. \thanks{Code is available in \url{https://github.com/datake/SinkhornDistRL}.}.

Knowledge transfer in cooperative multi-agent reinforcement learning (MARL) has drawn increasing attention in recent years. Unlike generalizing policies in single-agent tasks, it is more important to consider coordination knowledge than individual knowledge in multi-agent transfer learning. However, most of the existing methods only focus on knowledge transfer of the individual agent policy, which leads to coordination bias and finally affects the final performance in cooperative MARL. In this paper, we propose a level-adaptive MARL framework called “LA-QTransformer”, to realize the knowledge transfer on the coordination level via efficiently decomposing the agent coordination into multi-level coalition patterns for different agents. Compatible with centralized training with decentralized execution (CTDE) regime, LA-QTransformer utilizes the Level- Adaptive Transformer to generate suitable coalition patterns and then realizes the credit assignment for each agent. Besides, to deal with unexpected changes in the number of agents in the coordination transfer phase, we design a policy network called “Population invariant agent with Transformer (PIT)” to adapt dynamic observation and action space. We evaluate the LAQTransformer and PIT in the StarCraft II micro-management benchmark by comparing them with several state-of-the-art MARL baselines. The experimental results demonstrate the superiority of LA-QTransformer and PIT and verify the feasibility of coordination knowledge transfer.

We study Policy-extended Value Function Approximator (PeVFA) in Reinforcement Learning (RL), which extends conventional value function approximator (VFA) to take as input not only the state (and action) but also an explicit policy representation. Such an extension enables PeVFA to preserve values of multiple policies at the same time and brings an appealing characteristic, i.e., value generalization among policies. We formally analyze the value generalization under Generalized Policy Iteration (GPI). From theoretical and empirical lens, we show that generalized value estimates offered by PeVFA may have lower initial approximation error to true values of successive policies, which is expected to improve consecutive value approximation during GPI. Based on above clues, we introduce a new form of GPI with PeVFA which leverages the value generalization along policy improvement path. Moreover, we propose a representation learning framework for RL policy, providing several approaches to learn effective policy embeddings from policy network parameters or state-action pairs. In our experiments, we evaluate the efficacy of value generalization offered by PeVFA and policy representation learning in several OpenAI Gym continuous control tasks. For a representative instance of algorithm implementation, Proximal Policy Optimization (PPO) re-implemented under the paradigm of GPI with PeVFA achieves about 40% performance improvement on its vanilla counterpart in most environments.

Precise congestion prediction from a placement solution plays a crucial role in circuit placement. This work proposes the lattice hypergraph (LH-graph), a novel graph formulation for circuits, which preserves netlist data during the whole learning process, and enables the congestion information propagated geometrically and topologically. Based on the formulation, we further developed a heterogeneous graph neural network architecture LHNN, jointing the routing demand regression to support the congestion spot classification. LHNN constantly achieves more than 35% improvements compared with U-nets and Pix2Pix on the F1 score. We expect our work shall highlight essential procedures using machine learning for congestion prediction.

We propose Neuro-Symbolic Hierarchical Rule Induction, an efficient interpretable neuro-symbolic model, to solve Inductive Logic Programming (ILP) problems. In this model, which is built from a pre-defined set of meta-rules organized in a hierarchical structure, first-order rules are invented by learning embeddings to match facts and body predicates of a meta-rule. To instantiate, we specifically design an expressive set of generic meta-rules, and demonstrate they generate a consequent fragment of Horn clauses. As a differentiable model, HRI can be trained both via supervised learning and reinforcement learning. To converge to interpretable rules, we inject a controlled noise to avoid local optima and employ an interpretability-regularization term. We empirically validate our model on various tasks (ILP, visual genome, reinforcement learning) against relevant state-of-the-art methods, including traditional ILP methods and neuro-symbolic models.

In the recent years, the deep reinforcement learning community has achieved impressive success to tackle real-world challenges. In this work, we propose a novel deep reinforcement learning agent to perform floorplanning, one of the early stages of VLSI physical design. Traditional methods to solve floorplanning problem are intractable for large circuit netlists and impossible to learn from past experience. We adopt the domain knowledge of floorplanning representation and propose a learning-based method that directly predicts block id and location through an RL framework. The resulting solutions are platform-independent and can be converted into layout within $O(n)$ time. We encode the hypernet information in the circuit netlist in a one-to-one mapping through hypergraph neural networks. Furthermore, We deploy transformer-like action selection to allow for transferability and generalization across netlist circuits with different sizes and handle the large discrete action space. This allows the parameter space of our model to remain the same regardless of the number of blocks. Our RL agent is able to transfer previously learnt knowledge to quickly optimize a new design with different size and purpose. To our knowledge, this is the first work to select both id and block position with an entirely end-to-end learning-based framework that can generalize. Results on publicly available benchmarks of GSRC and MCNC demonstrate that our method can outperform the baselines while being able to generalize.

Algorithmic fairness has received increased attention in socially sensitive domains. While rich literature on mean fairness has been established, research on quantile fairness remains sparse but vital. To fulfill great needs and advocate the significance of quantile fairness, we propose a novel framework to learn a real-valued quantile function under the fairness requirement of Demographic Parity with respect to sensitive attributes, such as race or gender, and thereby derive a reliable fair prediction interval. Using optimal transport and functional synchronization techniques, we establish theoretical guarantees of distribution-free coverage and exact fairness for the induced prediction interval constructed by fair quantiles. A hands-on pipeline is provided to incorporate flexible quantile regressions with an efficient fairness adjustment post-processing algorithm. We demonstrate the superior empirical performance of this approach on several benchmark datasets. Our results show the model’s ability to uncover the mechanism underlying the fairness-accuracy trade-off in a wide range of societal and medical applications.

Deep reinforcement learning (DRL) algorithms have been demonstrated to be effective on a wide range of challenging decision making and control tasks. However, these methods typically suffer from severe action oscillations in particular in discrete action setting, which means that agents select different actions within consecutive steps even though states only slightly differ. This issue is often neglected since we usually evaluate the quality of a policy using cumulative rewards only. Action oscillation strongly affects the user experience and even causes serious potential security menace especially in real-world domains with the main concern of safety, such as autonomous driving. In this paper, we introduce Policy Inertia Controller (PIC) which serves as a generic plug-in framework to off-the-shelf DRL algorithms, to enable adaptive balance between the optimality and smoothness in a formal way. We propose Nested Policy Iteration as a general training algorithm for PIC-augmented policy which ensures monotonically non-decreasing updates.Further, we derive a practical DRL algorithm, namely Nested Soft Actor-Critic. Experiments on a collection of autonomous driving tasks and several Atari games suggest that our approach demonstrates substantial oscillation reduction than a range of commonly adopted baselines with almost no performance degradation.

Context, the embedding of previous collected trajectories, is a powerful construct for Meta-Reinforcement Learning (Meta-RL) algorithms. By conditioning on an effective context, Meta-RL policies can easily generalize to new tasks within a few adaptation steps. We argue that improving the quality of context involves answering two questions: 1. How to train a compact and sufficient encoder that can embed the task-specific information contained in prior trajectories? 2. How to collect informative trajectories of which the corresponding context reflects the specification of tasks? To this end, we propose a novel Meta-RL framework called CCM (Contrastive learning augmented Context-based Meta-RL). We first focus on the contrastive nature behind different tasks and leverage it to train a compact and sufficient context encoder. Further, we train a separate exploration policy and theoretically derive a new information-gain-based objective which aims to collect informative trajectories in a few steps. Empirically, we evaluate our approaches on common benchmarks as well as several complex sparse-reward environments. The experimental results show that CCM outperforms state-of-the-art algorithms by addressing previously mentioned problems respectively.

Value function is the central notion of Reinforcement Learning (RL). Value estimation, especially with function approximation, can be challenging since it involves the stochasticity of environmental dynamics and reward signals that can be sparse and delayed in some cases. A typical model-free RL algorithm usually estimates the values of a policy by Temporal Difference (TD) or Monte Carlo (MC) algorithms directly from rewards, without explicitly taking dynamics into consideration. In this paper, we propose Value Decomposition with Future Prediction (VDFP), providing an explicit two-step understanding of the value estimation process: 1) first foresee the latent future, 2) and then evaluate it. We analytically decompose the value function into a latent future dynamics part and a policy-independent trajectory return part, inducing a way to model latent dynamics and returns separately in value estimation. Further, we derive a practical deep RL algorithm, consisting of a convolutional model to learn compact trajectory representation from past experiences, a conditional variational auto-encoder to predict the latent future dynamics and a convex return model that evaluates trajectory representation. In experiments, we empirically demonstrate the effectiveness of our approach for both off-policy and on-policy RL in several OpenAI Gym continuous control tasks as well as a few challenging variants with delayed reward.

Partially observable Markov decision process (POMDP) is an extension to MDP. It handles the state uncertainty by specifying the probability of getting a particular observation given the current state. DESPOT is one of the most popular scalable online planning algorithms for POMDPs, which manages to significantly reduce the size of the decision tree while deriving a near-optimal policy by considering only $K$ scenarios. Nevertheless, there is a gap in action selection criteria between planning and execution in DESPOT. During the planning stage, it keeps choosing the action with the highest upper bound, whereas when the planning ends, the action with the highest lower bound is chosen for execution. Here, we propose LB-DESPOT to alleviate this issue, which utilizes the lower bound in selecting an action branch to expand. Empirically, our method has attained better performance than DESPOT and POMCP, which is another state-of-the-art, on several challenging POMDP benchmark tasks.

Meta-Reinforcement Learning (meta-RL) algorithms enable agents to adapt to new tasks from small amounts of exploration, based on the experience of similar tasks. Recent studies have pointed out that a good representation of a task is key to the success of off-policy context-based meta-RL. Inspired by contrastive methods in unsupervised representation learning, we propose a new method to learn the task representation based on the mutual information between transition tuples in a trajectory and the task embedding. We also propose a new estimation for task similarity based on Q-function, which can be used to form a constraint on the distribution of the encoded task variables, making the task encoder encode the task variables more effective on new tasks. Experiments on meta-RL tasks show that the newly proposed method outperforms existing meta-RL algorithms.

Presently with technology node scaling, an accurate prediction model at early design stages can significantly reduce the design cycle. Especially during logic synthesis, predicting cell congestion due to improper logic combination can reduce the burden of subsequent physical implementations. There have been attempts using Graph Neural Network (GNN) techniques to tackle congestion prediction during the logic synthesis stage. However, they require informative cell features to achieve reasonable performance since the core idea of GNNs is built on the message passing framework, which would be impractical at the early logic synthesis stage. To address this limitation, we propose a framework that can directly learn embeddings for the given netlist to enhance the quality of our node features. Popular random-walk based embedding methods such as Node2vec, LINE, and DeepWalk suffer from the issue of cross-graph alignment and poor generalization to unseen netlist graphs, yielding inferior performance and costing significant runtime. In our framework, we introduce a superior alternative to obtain node embeddings that can generalize across netlist graphs using matrix factorization methods. We propose an efficient mini-batch training method at the sub-graph level that can guarantee parallel training and satisfy the memory restriction for large-scale netlists. We present results utilizing open-source EDA tools such as DREAMPLACE and OPENROAD frameworks on a variety of openly available circuits. By combining the learned embedding on top of the netlist with the GNNs, our method improves prediction performance, generalizes to new circuit lines, and is efficient in training, potentially saving over 90% of runtime.

In this paper, a novel navigation learning method in continuous action space among crowds based on relational graph is proposed which can be directly deployed on differential-drive mobile robots without any change. More specifically, in order to increase generalization ability in crowd sizes, Graph Convolutional Network (GCN) is at first adopted to extract the relationships between robot and pedestrians. Then the relation features are further utilized as the inputs of the pedestrian state prediction network, the actor network, and the critic network. To efficiently and safely learn the navigation policy, all networks are pretrained through imitating ORCA which is a state-of-the-art algorithm in crowd navigation, and then a model-based reinforcement learning (RL) method which combines the model prediction and the clipped advantage-weighted regression is proposed to finetune the networks. Finally, simulation experiments are performed and it’s verified that the proposed learning method performs significantly better than ORCA and the other state-of-the-art RL methods.

For autonomous vehicles integrating onto road-ways with human traffic participants, it requires understanding and adapting to the participants’ intention by responding in predictable ways. This paper proposes a reinforcement learning based negotiation-aware motion planning framework, which adopts RL to adjust the driving style of the planner by dynamically modifying the prediction horizon length of the motion planner in real time adaptively. The framework models the interaction between the autonomous vehicle and other traffic participants as a Markov Decision Process. A temporal sequence of occupancy grid maps are taken as inputs for RL module to embed an implicit intention reasoning. Curriculum learning is employed to enhance the training efficiency and the robustness of the algorithm. We applied our method to narrow lane navigation in both simulation and real world to demonstrate that the proposed method outperforms the common alternative due to its advantage in alleviating the social dilemma problem with proper negotiation skills.

Shift neural networks reduce computation complexity by removing expensive multiplication operations and quantizing continuous weights into low-bit discrete values, which are fast and energy-efficient compared to conventional neural networks. However, existing shift networks are sensitive to the weight initialization and yield a degraded performance caused by vanishing gradient and weight sign freezing problem. To address these issues, we propose S$^3$ re-parameterization, a novel technique for training low-bit shift networks. Our method decomposes a discrete parameter in a sign-sparse-shift 3-fold manner. This way, it efficiently learns a low-bit network with weight dynamics similar to full-precision networks and insensitive to weight initialization. Our proposed training method pushes the boundaries of shift neural networks and shows 3-bit shift networks compete with their full-precision counterparts in terms of top-1 accuracy on ImageNet.

Model-based reinforcement learning aims to improve the sample efficiency of policy learning by modeling the dynamics of the environment. Recently, the latent dynamics model is further developed to enable fast planning in a compact space. It summarizes the high-dimensional experiences of an agent, which mimics the memory function of humans. Learning policies via imagination with the latent model shows great potential for solving complex tasks. However, only considering memories from the true experiences in the process of imagination could limit its advantages. Inspired by the memory prosthesis proposed by neuroscientists, we present a novel model-based reinforcement learning framework called Imagining with Derived Memory (IDM). It enables the agent to learn policy from enriched diverse imagination with prediction-reliability weight, thus improving sample efficiency and policy robustness. Experiments on various high-dimensional visual control tasks in the DMControl benchmark demonstrate that IDM outperforms previous state-of-the-art methods in terms of policy robustness and further improves the sample efficiency of the model-based method.

The partially observable Markov decision process (POMDP) provides a general framework for modeling an agent's decision process with state uncertainty, and online planning plays a pivotal role in solving it. A belief is a distribution of states representing state uncertainty. Methods for large-scale POMDP problems rely on the same idea of sampling both states and observations. That is, instead of exact belief updating, a collection of sampled states is used to approximate the belief; instead of considering all possible observations, only a set of sampled observations are considered. Inspired by this, we take one step further and propose an online planning algorithm, Adaptive Online Packing-guided Search (AdaOPS), to better approximate beliefs with adaptive particle filter technique and balance estimation bias and variance by fusing similar observation branches. Theoretically, our algorithm is guaranteed to find an $\epsilon$-optimal policy with a high probability given enough planning time under some mild assumptions. We evaluate our algorithm on several tricky POMDP domains, and it outperforms the state-of-the-art in all of them.

Transfer Learning has shown great potential to enhance single-agent Reinforcement Learning (RL) efficiency. Similarly, Multiagent RL (MARL) can also be accelerated if agents can share knowledge with each other. However, it remains a problem of how an agent should learn from other agents. In this paper, we propose a novel Multiagent Policy Transfer Framework (MAPTF) to improve MARL efficiency. MAPTF learns which agent's policy is the best to reuse for each agent and when to terminate it by modeling multiagent policy transfer as the option learning problem. Furthermore, in practice, the option module can only collect all agent's local experiences for update due to the partial observability of the environment. While in this setting, each agent's experience may be inconsistent with each other, which may cause the inaccuracy and oscillation of the option-value's estimation. Therefore, we propose a novel option learning algorithm, the successor representation option learning to solve it by decoupling the environment dynamics from rewards and learning the option-value under each agent's preference. MAPTF can be easily combined with existing deep RL and MARL approaches, and experimental results show it significantly boosts the performance of existing methods in both discrete and continuous state spaces.

Social psychology and real experiences show that cognitive consistency plays an important role to keep human society in order: if people have a more consistent cognition about their environments, they are more likely to achieve better cooperation. Meanwhile, only cognitive consistency within a neighborhood matters because humans only interact directly with their neighbors. Inspired by these observations, we take the first step to introduce neighborhood cognitive consistency (NCC) into multi-agent reinforcement learning (MARL). Our NCC design is quite general and can be easily combined with existing MARL methods. As examples, we propose neighborhood cognition consistent deep Q-learning and Actor-Critic to facilitate large-scale multi-agent cooperations. Extensive experiments on several challenging tasks (i.e., packet routing, wifi configuration and Google football player control) justify the superior performance of our methods compared with state-of-the-art MARL approaches.

Generative adversarial imitation learning (GAIL) has shown promising results by taking advantage of generative adversarial nets, especially in the field of robot learning. However, the requirement of isolated single modal demonstrations limits the scalability of the approach to real world scenarios such as autonomous vehicles' demand for a proper understanding of human drivers' behavior. In this paper, we propose a novel multi-modal GAIL framework, named Triple-GAIL, that is able to learn skill selection and imitation jointly from both expert demonstrations and continuously generated experiences with data augmentation purpose by introducing an auxiliary selector. We provide theoretical guarantees on the convergence to optima for both of the generator and the selector respectively. Experiments on real driver trajectories and real-time strategy game datasets demonstrate that Triple-GAIL can better fit multi-modal behaviors close to the demonstrators and outperforms state-of-the-art methods.

Transfer learning has shown great potential to accelerate Reinforcement Learning (RL) by leveraging prior knowledge from past learned policies of relevant tasks. Existing approaches either transfer previous knowledge by explicitly computing similarities between tasks or select appropriate source policies to provide guided explorations. However, how to directly optimize the target policy by alternatively utilizing knowledge from appropriate source policies without explicitly measuring the similarities is currently missing. In this paper, we propose a novel Policy Transfer Framework (PTF) by taking advantage of this idea. PTF learns when and which source policy is the best to reuse for the target policy and when to terminate it by modeling multi-policy transfer as an option learning problem. PTF can be easily combined with existing DRL methods and experimental results show it significantly accelerates RL and surpasses state-of-the-art policy transfer methods in terms of learning efficiency and final performance in both discrete and continuous action spaces.

Navigating through unsignalized intersections is one of the most challenging problems in urban environments for autonomous vehicles. Existing methods need to train specific policy models to deal with different tasks including going straight, turning left and turning right. In this paper we formulate intersection navigation as a multi-task reinforcement learning problem and propose a unified learning framework for all three navigation tasks at the intersections. We propose to represent multiple tasks with a unified four-dimensional vector, which elements mean a common sub-task and three specific target sub-tasks respectively. Meanwhile, we design a vectorized reward function combining with deep Q-networks (DQN) to learn to handle multiple intersection navigation tasks concurrently. We train the agent to navigate through intersections by adjusting the speed of the ego vehicle under given route. Experimental results in both simulation and realworld vehicle test demonstrate that the proposed multi-task DQN algorithm outperforms baselines for all three navigation tasks in several different intersection scenarios.

Power consumption is becoming more critical in modern integrated circuit (IC) designs and clock network is one of the major contributors for on-chip power. Resonant clock has been investigated as a potential solution to reduce the power consumption in clock network by recycling the energy with on-chip inductors. Most of the previous resonant clock work focuses on H-tree structures, while in this work, we propose a modeling and optimization method for the mesh structure, which suffers from the high power consumption more seriously than the tree structure. Closed-form expressions for the transfer function, skew, and power are derived. Based on these expressions, impacts of design factors, such as the buffer size, LC tank location, grid size, wire width, and the sparsity of buffers and LC tanks, are fully explored to make trade-offs among power, skew, and area, which can be used as design guidelines for top level resonant clock mesh in early design stages. The exploration is also extended to 3D ICs and different mesh structures are evaluated. Matlab-based implementation of the proposed simplified circuit model can achieve over 105 times speedup compared to SPICE-based simulation.

Through-silicon-via (TSV) could provide vertical connections among different dies in 3-D integrated circuits (3-D ICs), but the significant silicon area occupied by TSVs may bring great challenge to designers in 3-D clock tree synthesis (CTS), because only a few whitespace blocks can be used for clock TSV insertion after floorplan and placement are determined, specifically in the area-efficient 3-D IC designs. This paper proposes a whitespace-aware TSV arrangement algorithm in 3-D CTS, which mainly consists of three stages: sink preclustering, whitespace-aware 3-D method of means and medians (3-D-MMMs) topology generation, and deferred-merge embedding merging segment reconstruction. By leveraging the TSV-to-TSV coupling model, we also propose an efficient clock TSV arrangement method to alleviate the coupling effect of adjacent TSVs. Compared with the traditional 3-D-MMM-based CTS with TSV moving adjustment, the experimental results show that our proposed algorithm is more practical and efficient, achieving 49.2% reduction on the average skew and 1.9% reduction on the average power.

Three-dimensional (3D) integration technology has been proposed as a promising technology to provide small footprint, reduced wirelength, and the capability of heterogeneous integration. In particular, 3D IC is a good candidate to address the design issues in conventional analog/digital mixed-signal IC designs. In this work, we focus on modeling and analyzing the impacts of through silicon vias (TSVs) on mixed-signal ICs. Based on the analysis, a set of design methodologies for 3D mixed-signal ICs are proposed. The design methodologies are verified with a case study, in which a 12-bit successive approximation register analog-to-digital converter (SAR ADC) is re-designed by partitioning it into three stacked layers for 3D integration. The experimental results show that, compared to the traditional 2D counterpart, our 3D SAR ADC with optimized TSV placement can achieve significant area and power reduction, and performance improvement. Specifically, due to the isolation of substrate noise disturbance in our 3D design, the signal-to-noise-plus-distortion ratio (SNDR) is improved from 68.74 dB to 74.12 dB.

With the miniaturization of electronic devices, small-size but high-capacity power supply systems appear to be more and more important. A hybrid power source, which consists of a fuel cell (FC) and a rechargeable battery, has the advantages of long lifetime and good load-following capabilities. In this article, we propose the schematic of a hybrid power supply system that can be integrated on a chip compatible with present CMOS processes. For the on-chip, fuel-cell-based hybrid power system in wireless sensor node design, we propose a two steps optimization: (1) dynamic power management (DPM), and (2) adaptive fuel cell optimal power point tracking (AOPPT). Simulation results demonstrate that the on-chip FC-Bat hybrid power system can be used for wireless sensor nodes under different usage scenarios. Our proposed DPM method can achieve 12.9% more energy savings than the method without DPM. Meanwhile, implementing our AOPPT approach can save about 17% energy compared with the fixed architecture for the fuel cell system. For an on-chip power system with 1cm 2 area consumption, the wafer-level battery can power a typical sensor node for only about five months, while our on-chip hybrid power system will supply the same sensor node for two years steadily.

With the miniaturization of electronic devices, small size but high capacity power supply system appears to be more and more important. A hybrid power source, which consists of a fuel cell (FC) and a rechargeable battery, has the advantages of long lifetime and good load following capabilities. In this paper, we propose the schematic of a hybrid power supply system, that can be integrated on a chip compatible with present CMOS process. Besides, considering the problem of maximizing the on-chip fuel cell's lifetime, we propose a modified dynamic power management (DPM) algorithm for on-chip fuel cell based hybrid power system in wireless sensor node design. Taking the wireless sensor node powered by this hybrid power system as an example, we analyze the improvement of the FC-Bat hybrid power system. The simulation results demonstrate that the on-chip FC-Bat hybrid power system can be used for wireless sensor node under different usage scenarios. Meanwhile, for an on-chip power system with 1cm2area consumption, the wafer-level battery can power a typical sensor node for only about 5 months, while our on-chip hybrid power system will supply the same sensor node for 2 years steadily.