United States

Binary Spiking Neural Networks (BSNNs) inherit the event-driven paradigm of SNNs, while also adopting the reduced storage burden of binarization techniques. These distinct advantages grant BSNNs lightweight and energy-efficient characteristics, rendering them ideal for deployment on resource-constrained edge devices. However, due to the binary synaptic weights and non-differentiable spike function, effectively training BSNNs remains an open question. In this paper, we conduct an in-depth analysis of the challenge in BSNNs learning: frequent weight sign flipping problem. To overcome this challenge, we propose an Adaptive Gradient Modulation Mechanism (AGMM), which is designed to reduce the frequency of weight sign flipping by adaptively adjusting the gradients in the learning process. The proposed AGMM improves performance while achieving higher energy efficiency by reducing the firing rate. We validate AGMM on both static and neuromorphic datasets, and results indicate that it achieves state-of-the-art results among BSNNs. This work substantially narrows the performance gap between BSNNs and full-precision SNNs, while leveraging low storage requirements and enhancing SNNs&#39; inherent energy efficiency, making them feasible for resource-constrained environments.

AAAI 2025

Towards Accurate Binary Spiking Neural Networks: Learning with Adaptive Gradient Modulation Mechanism

Binary Spiking Neural Networks (BSNNs) inherit the event-driven paradigm of SNNs, while also adopting the reduced storage burden of binarization techniques. These distinct advantages grant BSNNs lightweight and energy-efficient characteristics, rendering them ideal for deployment on resource-constrained edge devices. However, due to the binary synaptic weights and non-differentiable spike function, effectively training BSNNs remains an open question. In this paper, we conduct an in-depth analysis of the challenge in BSNNs learning: frequent weight sign flipping problem. To overcome this challenge, we propose an Adaptive Gradient Modulation Mechanism (AGMM), which is designed to reduce the frequency of weight sign flipping by adaptively adjusting the gradients in the learning process. The proposed AGMM improves performance while achieving higher energy efficiency by reducing the firing rate. We validate AGMM on both static and neuromorphic datasets, and results indicate that it achieves state-of-the-art results among BSNNs. This work substantially narrows the performance gap between BSNNs and full-precision SNNs, while leveraging low storage requirements and enhancing SNNs' inherent energy efficiency, making them feasible for resource-constrained environments.

technical paper

We are pleased to announce the Thirty-Ninth AAAI Conference on Artificial Intelligence (AAAI-25), which will be held in Philadelphia, Pennsylvania at the Pennsylvania Convention Center from February 25 to March 4, 2025.

The purpose of the AAAI conference series is to promote research in Artificial Intelligence (AI) and foster scientific exchange between researchers, practitioners, scientists, students, and engineers across the entirety of AI and its affiliated disciplines. AAAI-25 will feature technical paper presentations, special tracks, invited speakers, workshops, tutorials, poster sessions, senior member presentations, competitions, and exhibit programs, and a range of other activities to be announced.

### [Invited Speakers](https://aaai.org/conference/aaai/aaai-25/aaai-25-invited-speakers/)

Register [here](https://aaai.org/conference/aaai/aaai-25/registration/)

The purpose of the AAAI conference series is to promote research in Artificial Intelligence (AI) and foster scientific exchange between researchers, practitioners, scientists, students, and engineers across the entirety of AI and its affiliated disciplines. AAAI-25 will feature technical paper presentations, special tracks, invited speakers, workshops, tutorials, poster sessions, senior member presentations, competitions, and exhibit programs, and a range of other activities to be announced.



Learning a perception and reasoning module for robotic assistants to plan steps to perform complex tasks based on natural language instructions often requires large free-form language annotations, especially for short high-level instructions. To reduce the cost of annotation, large language models (LLMs) are used as a planner with few data. However, when elaborating the steps, even the state-of-the-art planner that uses LLMs mostly relies on linguistic common sense, often neglecting the status of the environment at command reception, resulting in inappropriate plans. To generate plans grounded in the environment, we propose FLARE (Few-shot Language with environmental Adaptive Replanning Embodied agent), which improves task planning using both language command and environmental perception. As language instructions often contain ambiguities or incorrect expressions, we additionally propose to correct the mistakes using visual cues from the agent. The proposed scheme allows us to use a few language pairs thanks to the visual cues and significantly outperforms state-of-the-art approaches (by twice the success rate in unseen environments of the ALFRED benchmark: $16.42$\% $\to$ $40.88$\%).

Multi-Modal Grounded Planning and Efficient Replanning for Learning Embodied Agents with a Few Examples

We study the problem of modeling a non-linear dynamical system when given a time series by deriving equations directly from the data. Despite the fact that time series data are given as input, models for dynamics and estimation algorithms that incorporate long-term temporal dependencies are largely absent from existing studies. In this paper, we introduce a latent state to allow time-dependent modeling and formulate this problem as a dynamics estimation problem in latent states. We face multiple technical challenges, including (1) modeling latent non-linear dynamics and (2) solving circular dependencies caused by the presence of latent states. To tackle these challenging problems, we propose a new method, Latent Non-Linear equation modeling (LaNoLem), that can model a latent non-linear dynamical system and a novel alternating minimization algorithm for effectively estimating latent states and model parameters. In addition, we introduce criteria to control model complexity without human intervention. Compared with the state-of-the-art model, LaNoLem achieves competitive performance for estimating dynamics while outperforming other methods in prediction.

Modeling Latent Non-Linear Dynamical System over Time Series

Visual Language Models such as CLIP excel in image recognition due to extensive image-text pre-training. However, applying the CLIP inference in zero-shot classification, particularly for medical image diagnosis, faces challenges due to: 1) {the inadequacy of representing image classes solely with single category names}; 2) {the modal gap between the visual and text spaces generated by CLIP encoders}. Despite attempts to enrich disease descriptions with large language models, the lack of class-specific knowledge often leads to poor performance. In addition, empirical evidence suggests that existing proxy learning methods for zero-shot image classification on natural image datasets exhibit instability when applied to medical datasets. 
To tackle these challenges, we introduce the Knowledge Proxy Learning (KPL) to mine knowledge from CLIP. 
KPL is designed to leverage CLIP's multimodal understandings for medical image classification through {Text Proxy Optimization} and {Multimodal Proxy Learning}.
Specifically, KPL retrieves image-relevant knowledge descriptions from the constructed knowledge-enhanced base to enrich semantic text proxies.
It then harnesses input images and these descriptions, encoded via CLIP, to stably generate multimodal proxies that boost the zero-shot classification performance.
Extensive experiments conducted on both medical and natural image datasets demonstrate that KPL enables effective zero-shot image classification, outperforming all baselines. These findings highlight the great potential in this paradigm of mining knowledge from CLIP for medical image classification and broader areas. Code will be publicly available.

KPL: Training-Free Medical Knowledge Mining of Vision-Language Models

We investigate the problem of checking the consistency of qualitative preferences expressed in CP-theory based languages. The problem is
PSPACE-Complete even when the preferences are locally consistent or the preference variables have binary domain. We present a new sufficient condition for consistency of preferences and show that the condition can be verified in polynomial time in settings of practical relevance (locally consistent or binary domain preference variables). We show how the resulting sufficient condition can be used to efficiently identify of a subset of outcomes that are non-dominated with respect to a set of qualitative preferences.

Checking Consistency of CP-Theory Preferences in Polynomial Time

Functional decomposition is the process of breaking down a function $f$ into a composition $f=g(f_1,\dots,f_k)$ of simpler functions $f_1,\dots,f_k$ belonging to some class $\mathcal{F}$. This fundamental notion can be used to model applications arising in a wide variety of contexts, ranging from machine learning to formal language theory. 
In this work, we study functional decomposition by leveraging on the notion of functional reconfiguration. In this setting, constraints are imposed not only on the factor functions $f_1,\dots,f_k$ but also on the intermediate functions arising during the composition process. 

We introduce a symbolic framework to address functional reconfiguration and decomposition problems. In our framework, functions arising during the reconfiguration process are represented symbolically, using ordered binary decision diagrams (OBDDs). The function $g$  used to specify the reconfiguration process is represented by a Boolean circuit $C$. Finally, the function class $\mathcal{F}$ is represented by a second-order finite automaton $\mathcal{A}$. Our main result states that functional reconfiguration, and hence functional decomposition, can be solved in fixed-parameter linear time when parameterized by the width of the input OBDD, by structural parameters associated with the reconfiguration circuit $C$, and by the size of the second-order finite automaton $\mathcal{A}$.

Symbolic Functional Decomposition: A Reconfiguration Approach

Few-shot Named Entity Recognition (NER) spotlights the tag of novel entity types in data-limited scenarios or lower-resource settings. Advances with Pre-trained Language Models (PLMs), including BERT, GPT, and their variants, have driven tremendous strategies to leverage context-dependent representations and exploit predefined relational cues, yielding significant gains in witnessing unseen entities. Nevertheless, a fundamental issue exists in prior efforts regarding their susceptibility to adversarial attacks in the intricate semantic environment. This vulnerability undermines the robustness of semantic representations, exacerbating the challenge of accurate entity identification, especially when transitioning across domains. To this end, we propose an $A$dversity-aware $A$ugment $L$earning ($AAL$) solution for the few-shot NER task, dedicated to retrieving and reinforcing entity prototypes resilient to adversarial inference, thereby enhancing cross-domain semantic coherence. In particular, AAL employs a two-stage paradigm consisting of training and fine-tuning. The process initiates with augmentation learning by leveraging two kinds of prompt learning schemes, then identifies prototypes under the guidance of a variational manner. Furthermore, we devise a domain-oriented prototype refinement to optimize prototype learning under conditions of uncertainty attack, facilitating the effective transfer of common knowledge from source to target domains. The experimental results, encompassing the few-shot NER datasets under both certainty and uncertainty conditions, affirm the superiority of the proposed AAL over several representative baselines, particularly its capability against adversarial attacks.

Adversity-aware Few-shot Named Entity Recognition via Augmentation Learning

Large Language Models (LLMs) have demonstrated potential in Vision-and-Language Navigation (VLN) tasks, yet current applications face challenges. While LLMs excel in general conversation scenarios, they struggle with specialized navigation tasks, yielding sub-optimal performance compared to specialized VLN models. We introduce FLAME (FLAMingo-Architectured Embodied Agent), a novel multimodal LLM-based agent and architecture designed for urban VLN tasks that efficiently handles multiple observations. Our approach implements a three-phase tuning technique for effective adaptation to navigation tasks, including single perception tuning for street view description, multiple perception tuning for trajectory summarization, and end-to-end training on VLN datasets. The augmented datasets are synthesized automatically. Experimental results demonstrate FLAME's superiority over existing methods, surpassing state-of-the-art methods by a 7.3\% increase in task completion rate on Touchdown dataset. This work showcases the potential of multimodal LLMs (MLLMs) in complex navigation tasks, representing an advancement towards practical applications of MLLMs in embodied AI. Codes will be made publicly available upon acceptance.

FLAME: Learning to Navigate with Multimodal LLM in Urban Environments

Evaluating deep reinforcement learning (DRL) agents against targeted behavior attacks is critical for assessing their robustness. These attacks aim to manipulate the victim into specific behaviors that align with the attacker’s objectives, often bypassing traditional reward-based defenses. Prior methods have primarily focused on reducing cumulative rewards; however, rewards are typically too generic to capture complex safety requirements effectively. As a result, focusing solely on reward reduction can lead to suboptimal attack strategies, particularly in safety-critical scenarios where more precise behavior manipulation is needed. To address these challenges, we propose RAT, a method designed for universal, targeted behavior attacks. RAT trains an intention policy that is explicitly aligned with human preferences, serving as a precise behavioral target for the adversary. Concurrently, an adversary manipulates the victim's policy to follow this target behavior. To enhance the effectiveness of these attacks, RAT dynamically adjusts the state occupancy measure within the replay buffer, allowing for more controlled and effective behavior manipulation. Our empirical results on robotic simulation tasks demonstrate that RAT outperforms existing adversarial attack algorithms in inducing specific behaviors. Additionally, RAT shows promise in improving agent robustness, leading to more resilient policies. We further validate RAT by guiding Decision Transformer agents to adopt behaviors aligned with human preferences in various MuJoCo tasks, demonstrating its effectiveness across diverse tasks.

RAT: Adversarial Attacks on Deep Reinforcement Agents for Targeted Behaviors

In eXplainable Constraint Solving (XCS), it is common to extract a Minimal Unsatisfiable Subset (MUS) from a set of unsatisfiable constraints. They help explain to a user why a constraint specification does not admit a solution. Finding MUSes can be computationally expensive for highly symmetric problems, as many combinations of constraints need to be considered. In the traditional context of solving satisfaction problems, symmetry has been well studied, and effective ways to detect and exploit symmetries during the search exist. However, in the setting of finding MUSes of unsatisfiable constraint programs, symmetries are understudied. In this paper, we take inspiration from existing symmetry-handling techniques and adapt well-known MUS-computation methods to exploit symmetries in the specification, speeding-up overall computation time. Our results display a significant reduction of runtime for our adapted algorithms compared to the baseline when symmetries are present in the problem.

Exploiting Symmetries in MUS Computation

Adding imperceptible watermarks to artwork images, such as paintings and photographs, can effectively safeguard the copyright of these images without compromising their usability. However, existing blind watermarking techniques encounter two major challenges in addressing this task: imperceptibility and robustness, particularly when subjected to various noise attacks. In this paper, we propose a blind watermarking method for artwork image copyright protection, IWRN, which can ensure both the Imperceptibility of the Watermark and Robustness against Noise attacks. For imperceptibility, we design a Learnable Wavelet Network (LWN) to adaptively embed the watermark into the high-frequency region  where the watermark has better invisibility.
    For robustness, we establish a Deform-Attention based Invertible Neural Network (DA-INN) with a decoding optimization, which offers the advantage of computational reversion, and combines the deform-attention mechanism and decoding optimization to enhance the model's resistance against noises. Additionally, we design a Joint Contrast Learning (JCL) mechanism  to improve imperceptibility and robustness simultaneously. Experiments show that our IWRN outperforms other state-of-the-art blind watermarking methods, achieves an average performance of 41.55 PSNR and 99.57% accuracy on the Coco2017, Wikiart, and Div2k datasets when facing 12 kinds of noise attacks. Our code will soon be available.

Premium content

Next from AAAI 2025

Multi-Modal Grounded Planning and Efficient Replanning for Learning Embodied Agents with a Few Examples

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES