StARformer: Transformer With State-Action-Reward Representations for Robot Learning.

Reinforcement Learning (RL) can be considered as a sequence modeling task, where an agent employs a sequence of past state-action-reward experiences to predict a sequence of future actions. In this work, we propose State-Action-Reward Transformer (StARformer), a Transformer architecture for robot learning with image inputs, which explicitly models short-term state-action-reward representations (StAR-representations), essentially introducing a Markovian-like inductive bias to improve long-term modeling. StARformer first extracts StAR-representations using self-attending patches of image states, action, and reward tokens within a short temporal window. These StAR-representations are combined with pure image state representations, extracted as convolutional features, to perform self-attention over the whole sequence. Our experimental results show that StARformer outperforms the state-of-the-art Transformer-based method on image-based Atari and DeepMind Control Suite benchmarks, under both offline-RL and imitation learning settings. We find that models can benefit from our combination of patch-wise and convolutional image embeddings. StARformer is also more compliant with longer sequences of inputs than the baseline method. Finally, we demonstrate how StARformer can be successfully applied to a real-world robot imitation learning setting via a human-following task.

Complete mitochondrial genome of Conuslischkeanus Weinkauff, 1875 (Neogastropoda, Conidae) and phylogenetic implications of the evolutionary diversification of dietary types of Conus species.

The family Conidae, commonly known as cone snails, is one of the most intriguing gastropod groups owing to their diverse array of feeding behaviors (diets) and toxin peptides (conotoxins). Conuslischkeanus Weinkauff, 1875 is a worm-hunting species widely distributed from Africa to the Northwest Pacific. In this study, we report the mitochondrial genome sequence of C.lischkeanus and inferred its phylogenetic relationship with other Conus species. Its mitochondrial genome is a circular DNA molecule (16,120 bp in size) composed of 37 genes: 13 protein-coding genes (PCGs), 22 transfer RNA genes, and two ribosomal RNA genes. Phylogenetic analyses of concatenated nucleotide sequences of 13 PCGs and two ribosomal RNA genes showed that C.lischkeanus belongs to the subgenus Lividoconus group, which is grouped with species of the subgenus Virgiconus, and a member of the largest assemblage of worm-hunting (vermivorous) species at the most basal position in this group. Mitochondrial genome phylogeny supports the previous hypothesis that the ancestral diet of cone snails was worm-hunting, and that other dietary types (molluscivous or piscivorous) have secondarily evolved multiple times from different origins. This new, complete mitochondrial genome information provides valuable insights into the mitochondrial genome diversity and molecular phylogeny of Conus species.

Spherically Stratified Point Projection: Feature Image Generation for Object Classification Using 3D LiDAR Data.

Three-dimensional point clouds have been utilized and studied for the classification of objects at the environmental level. While most existing studies, such as those in the field of computer vision, have detected object type from the perspective of sensors, this study developed a specialized strategy for object classification using LiDAR data points on the surface of the object. We propose a method for generating a spherically stratified point projection (sP2) feature image that can be applied to existing image-classification networks by performing pointwise classification based on a 3D point cloud using only LiDAR sensors data. The sP2's main engine performs image generation through spherical stratification, evidence collection, and channel integration. Spherical stratification categorizes neighboring points into three layers according to distance ranges. Evidence collection calculates the occupancy probability based on Bayes' rule to project 3D points onto a two-dimensional surface corresponding to each stratified layer. Channel integration generates sP2 RGB images with three evidence values representing short, medium, and long distances. Finally, the sP2 images are used as a trainable source for classifying the points into predefined semantic labels. Experimental results indicated the effectiveness of the proposed sP2 in classifying feature images generated using the LeNet architecture.

Phylogeography of Mytilisepta virgata (Mytilidae: Bivalvia) in the northwestern Pacific: Cryptic mitochondrial lineages and mito-nuclear discordance.

The purplish bifurcate mussel Mytilisepta virgata is widely distributed and represents one of the major components of the intertidal community in the northwestern Pacific (NWP). Here, we characterized population genetic structure of NWP populations throughout nearly their whole distribution range using both mitochondrial (mtDNA cox1) and nuclear (ITS1) markers. Population genetic analyses for mtDNA cox 1 sequences revealed two monophyletic lineages (i.e., southern and northern lineages) geographically distributed according to the two different surface water temperature zones in the NWP. The timing of the lineage split is estimated at the Pliocene- mid-Pleistocene (5.49-1.61 Mya), which is consistent with the timing of the historical isolation of the East Sea/Sea of Japan from the South and East China Seas due to sea level decline during glacial cycles. Historical sea level fluctuation during the Pliocene-Pleistocene and subsequent adaptation of mussels to different surface water temperature zones may have contributed to shaping the contemporary genetic diversity and deep divergence of the two mitochondrial lineages. In contrast to mtDNA sequences, a clear lineage split between the two mitochondrial lineages was not found in ITS1 sequences, which showed a star-like structure composed of a mixture of southern and northern mitochondrial lineages. Possible reasons for this type of mito-nuclear discordance include stochastic divergence in the coalescent processes of the two molecular markers, or balancing selection under different marine environments. Cryptic speciation cannot be ruled out from these results, and future work using genomic analyses is required to address whether the thermal physiology of these mussels corresponds to the deep divergence of their mitochondrial genes and to test for the existence of morphologically indistinguishable but genetically separate cryptic species.

Life-history features and oceanography drive phylogeographic patterns of the chiton Acanthochitona cf. rubrolineata (Lischke, 1873) in the northwestern Pacific.

Chitons are a group of marine mollusks (class Polyplacophora) characterized by having eight articulating shell plates on their dorsal body surface. They represent suitable materials for studying the spatiotemporal processes that underlie population differentiation and speciation in ocean environments. Here we performed population genetic analyses on the northwestern Pacific chiton Acanthochitona cf. rubrolineata (Lischke, 1873) using two mitochondrial gene fragments (COI and 16S) from 180 individuals sampled from 11 populations among the coastal waters of Korea, Japan, and China. The phylogenetic network uncovered a reticulated relationship with several sub-haplogroups for all A. cf. rubrolineata haplotypes. SAMOVA analyses suggested the best grouping occurred at three groups (ΦCT = 0.151, P < 0.0001), which geographically corresponds to hydrographic discontinuity among the coastal regions of Korea, Japan, and China. The assumed limited dispersal ability of A. cf. rubrolineata, coupled with northeasterly flowing, trifurcate warm currents, might have contributed to the genetic differentiation among the three groups. Meanwhile, a high level of within-group genetic homogeneity was detected, indicating extensive coastal currents might facilitate gene flow among the populations within each group. Bayesian skyline plots demonstrated significant population expansion after the Last Glacial Period (110-25 thousand years ago) for all studied populations except the Japan group. Together these results suggest that the present-day phylogeographic patterns of A. cf. rubrolineata are strongly affected by the interplay of historical and/or contemporary oceanography and species-specific life-history features.

The mitochondrial genome of Acrobeloides varius (Cephalobomorpha) confirms non-monophyly of Tylenchina (Nematoda).

The infraorder Cephalobomorpha is a diverse and ecologically important nematode group found in almost all terrestrial environments. In a recent nematode classification system based on SSU rDNA, Cephalobomorpha was classified within the suborder Tylenchina with Panagrolaimomorpha, Tylenchomorpha and Drilonematomorpha. However, phylogenetic relationships among species within Tylenchina are not always consistent, and the phylogenetic position of Cephalobomorpha is still uncertain. In this study, in order to examine phylogenetic relationships of Cephalobomorpha with other nematode groups, we determined the complete mitochondrial genome sequence of Acrobeloides varius, the first sequenced representative of Cephalobomorpha, and used this sequence for phylogenetic analyses along with 101 other nematode species. Phylogenetic analyses using amino acid and nucleotide sequence data of 12 protein-coding genes strongly support a sister relationship between the two cephalobomorpha species A. varius and Acrobeles complexus (represented by a partial mt genome sequence). In this mitochondrial genome phylogeny, Cephalobomorpha was sister to all chromadorean species (excluding Plectus acuminatus of Plectida) and separated from Panagrolaimomorpha and Tylenchomorpha, rendering Tylenchina non-monophyletic. Mitochondrial gene order among Tylenchina species is not conserved, and gene clusters shared between A. varius and A. complexus are very limited. Results from phylogenetic analysis and gene order comparison confirms Tylenchina is not monophyletic. To better understand phylogenetic relationships among Tylenchina members, additional mitochondrial genome information is needed from underrepresented taxa representing Panagrolaimomorpha and Cephalobomorpha.

A mitochondrial genome phylogeny of Mytilidae (Bivalvia: Mytilida).

The family Mytilidae is a family of bivalve mussels that are distributed worldwide in diverse marine habitats. Within the family, classification systems and phylogenetic relationships among subfamilies remain not yet fully resolved. In this study, we newly determined 9 mitochondrial genome sequences from 7 subfamilies: Bathymodiolus thermophilus (Bathymodiolinae), Modiolus nipponicus (Modiolinae), Lithophaga curta (the first representative of Lithophaginae), Brachidontes mutabilis (Brachidontinae), Mytilisepta virgata (Brachidontinae), Mytilisepta keenae (Brachidontinae), Crenomytilus grayanus (Mytilinae), Gregariella coralliophaga (Crenellinae), and Septifer bilocularis (the first representative of Septiferinae). Phylogenetic trees using maximum likelihood and Bayesian inference methods for 28 mitochondrial genomes (including 19 previously published sequences) showed two major clades with high support values: Clade 1 ((Bathymodiolinae + Modiolinae) + (Lithophaginae + Limnoperninae)) and Clade 2 (((Mytilinae + Crenellinae) + Septiferinae) + Brachidontinae). The position of the genus Lithophaga (representing Lithophaginae) differed from a previously published molecular phylogeny. Divergence time analysis with a molecular clock indicated that lineage splitting among the major subfamilies of Mytilidae (including the habitat transition from marine to freshwater environments by ancestral Limnoperninae) occurred in the Mesozoic period, coinciding with high diversification rates of marine fauna during that time. This is the first mitochondrial genome-based phylogenetic study of the Mytilidae that covers nearly all subfamily members, excluding the subfamily Dacrydiinae.

A revision of the genus Littorina (Mollusca: Gastropoda) in Korea.

Littorina Férussac, 1822 is an abundant genus of small gastropods found in the upper littoral zone of rocky seashores worldwide. Although ecologically important, shell-based species identification in this genus is challenging due to phenotypic variation in shell morphology and lack of diagnostic characters among morphologically similar species. In this study, we revised the taxonomy of Korean Littorina species using morphological characters (shell and radula) and cox1 mitochondrial DNA sequences for three Korean species: L. brevicula, L. sitkana, and L. horikawai. Results suggest that L. sitkana was erroneously reported as L. kasatka in a previous study. A new record for Littorina horikawai (Matsubayashi & Habe in Habe, 1979), previously unknown from Korea, is described, which can be distinguished from L. sitkana by the presence of alternating white and brown spiral ribs on each whorl. Comparison of the mtDNA cox1 gene sequences shows very low intraspecific variation even between geographically distant populations. A phylogenetic tree supports a close relationship between L. horikawai and L. sitkana, consistent with earlier phylogenetic studies.

The Korean Version of the University of California San Diego Performance-based Skills Assessment: Reliability and Validity.

OBJECTIVE: The study's aim was to develop and standardize a Korean version of the University of California San Diego Performance-based Skills Assessment (K-UPSA), which is used to evaluate the daily living function of patients with schizophrenia. METHODS: Study participants were 78 patients with schizophrenia and 27 demographically matched healthy controls. We evaluated the clinical states and cognitive functions to verify K-UPSA's reliability and validity. For clinical states, the Positive and Negative Syndrome Scale, Clinical Global Impression-Schizophrenia scale, and Social and Occupational Functioning Assessment Scale and Schizophrenia Quality of Life Scale-fourth revision were used. The Schizophrenia Cognition Rating Scale, Short-form of Korean-Wechsler Adult Intelligence Scale, and Wisconsin Card Sorting Test were used to assess cognitive function. RESULTS: The K-UPSA had statistically significant reliability and validity. The K-UPSA has high internal consistency (Cronbach's alpha, 0.837) and test-retest reliability (intra-class correlation coefficient, 0.381-0.792; p＜0.001). The K-UPSA had significant discriminant validity (p＜0.001). Significant correlations between the K-UPSA's scores and most of the scales and tests listed above demonstrated K-UPSA's concurrent validity (p＜0.001). CONCLUSION: The K-UPSA is useful to evaluate the daily living function in Korean patients with schizophrenia.

The F type mitochondrial genome of hard-shelled mussel: Mytilus coruscus (Mytiloida, Mytilidae).

We determined the complete mitochondrial genome of the female hard-shelled mussel Mytilus coruscus (Gould, 1869) (F type). The F type genome is composed of 16,642 bp in length including 12 protein coding genes (PCGs), 2 rRNA and 23 tRNA with the same gene content and order as the other Mytilus species which is characterized by the absence of the ATPase8 gene and addition of tRNA(Met) (AUA). The nucleotide composition of the genome shows that the percentage of A+T (63.2%) is higher than those of M. edulis complex species (M. edulis, M. galloprovincialis, M. trossulus) but lower than that of M. californianus. The F type mitochondrial genome of M. coruscus will provide useful information on the evolutionary history of the Mytilus species in the family Mitilidae.

The complete mitochondrial genome of the marine polychaete: Perinereis aibuhitensis (Phyllodocida, Nereididae).

We determined the complete mitochondrial genome sequence of the marine polychaete Perinereis aibuhitensis (Grube, 1878) (Phyllodocida, Nereididae), a dominant species in the mudflat of the West Pacific and Indian Ocean. The complete genome of P. aibuhitensis is 15,852 bp in size including 13 protein coding genes (PCGs), 2 rRNA and 22 tRNA with the same gene order and structure as those of other Nereididae species. The nucleotide composition is 29.5% A, 21.2% C, 14.1% G, 35.2% T, showing a high content of A+T with G being used least in the third codon position (6.7%). All PCGs use ATG as the start codon while for the stop codon COI and ND1 use incomplete codon of T. The mitogenome sequence of P. aibuhitensis is second to that of P. nuntia known in the genus Perinereis, which will provide useful information for understanding evolutionary history of the genus Perinereis within the family Nereididae.